JPWO2006033227A1 - Image encoding device - Google Patents

Abstract

An image encoding device (100) that compresses and encodes a plurality of moving images corresponds to an arbitrary number of moving image inputs, and acquires a number acquisition unit (110) that acquires the number of moving images to be encoded, and an encoding target 1 Alternatively, a moving image acquisition unit (120) that acquires a plurality of moving images, and a processing method of an encoding process that affects the calculation amount of the encoding process, for example, a processing method related to a reference image frame number upper limit, a motion vector search range, According to the number acquired by the number acquisition unit (110), the processing method specifying unit (130) that specifies the amount of calculation to decrease as the number increases, and the moving image acquired by the moving image acquisition unit (120) An encoding unit (140) that performs encoding processing as a target and performs encoding processing in a time-sharing manner for each moving image if there are a plurality of acquired videos, and the encoding unit (140) includes a processing method Specific part (1 Encoding process is performed in the processing method specified by 0).

Description

  The present invention relates to a moving image compression encoding technique, and more particularly, to an image encoding device that compresses and encodes a plurality of moving images.

  Conventionally, as a standard of compression encoding (hereinafter simply referred to as “encoding”) used when moving images are recorded or recorded on a recording medium, the MPEG (Moving Picture Experts Group) standard, for example, the MPEG2 video standard (ISO) / IEC18818-2), MPEG4 visual standard (ISO / IEC14496-2) or MPEG4AVC standard (Moving Picture Experts Group phase 4 Advanced Video Coding, ISO14496-10) is known. Then, an image encoding apparatus that performs encoding according to these standards has been developed.

In addition, as an image encoding device that encodes a plurality of moving images in parallel and transmits them over a network, it has a plurality of systems of encoding processing means such as frequency converters, quantizers, variable length encoders, etc. Therefore, an image encoding device that controls the output bit rate of each encoded moving image and keeps the overall bit rate constant has been proposed (see Japanese Patent Application Laid-Open No. 2003-228688).
This conventional image encoding device that encodes a plurality of moving images in parallel converts an input moving image from a spatial domain to a frequency domain with a frequency converter, and further according to a quantization coefficient with a quantizer. Quantization is performed and converted into a variable-length code by a variable-length encoder and output. A quantization coefficient is calculated according to a value obtained by summing up the complexity of the first moving image and the second moving image that are input. The output bit rate is controlled by determining and performing quantization.
JP 9-512417 A

The above-described conventional image encoding apparatus that encodes a plurality of moving images in parallel has the same number of encoding means as the number of input moving images, so that the number of input moving images is constant. It is suitable for the form.
By the way, as a usage mode of the image encoding device, there may be a usage mode in which the number of input moving images is not always constant. For example, an image encoding device is incorporated in a television broadcast receiving device or the like, and receives several videos arbitrarily selected by the user from videos broadcast on a plurality of channels in parallel. It is assumed that the image encoding device is used in such a manner that the data is compressed and stored in a recording medium such as an optical disk by performing the conversion processing.

  In order to cope with a usage mode in which the number of input moving images is not always constant, the above-described conventional image encoding device previously includes encoding means corresponding to the maximum number of moving images that allow input. It will be built in. However, if the user inputs a smaller number of moving pictures than the maximum number, some encoding means are not used and are extra. This is not preferable from the viewpoint of efficient use of internal circuits constituting the image coding apparatus.

  Therefore, the present invention has been made in view of such problems, and an image encoding apparatus that performs an encoding process in response to an input of an arbitrary number of moving images by effectively utilizing a finite number of encoding means. The purpose is to provide.

  In order to solve the above problems, an image encoding device according to the present invention is an image encoding device that performs encoding processing including motion prediction processing in order to compress a moving image, and the number of moving images to be encoded The number acquisition means for acquiring the number, the moving picture acquisition means for acquiring one or more moving pictures to be encoded, and the motion prediction processing method that affects the amount of calculation of the motion prediction processing are acquired by the number acquisition means. In accordance with the number of videos, the motion prediction processing method specifying means for specifying the calculation amount to decrease as the number increases, and the moving picture acquired by the moving picture acquisition means are subjected to encoding processing, and the acquired moving picture Encoding means for performing time-division encoding processing for each moving image if the number is a plurality, and the encoding means is specified by the motion prediction processing method specifying means as part of the encoding processing It characterized by having a motion prediction unit performing motion prediction processing by the processing method.

Here, the calculation amount means the processing amount of the encoding process required for one moving image for the circuit responsible for the encoding process, and a large calculation amount means that it takes time to complete the process. means.
In addition, when the circuit responsible for the encoding process realizes the encoding process by a processor that executes software, the amount of calculation means the number of instructions or the instruction execution time of the arithmetic instruction for realizing the encoding process. .

With the above-described configuration, the image encoding device according to the present invention effectively uses the encoding means by adjusting the amount of calculation required for the encoding process according to the number of moving images to be encoded in parallel, An encoding process corresponding to input of an arbitrary number of moving images can be performed.
That is, the image encoding device according to the present invention encodes, for example, one moving image in order to reduce the amount of calculation required for motion prediction processing per moving image when the number of moving images to be encoded is large. The amount of computation required and the total amount of computation required to encode a plurality of moving images can be suppressed to a certain level. It should be noted that the motion prediction process requires a large amount of calculation in the encoding process compared to other processes in practice, and suppressing the calculation amount of the motion prediction process greatly reduces the calculation amount of the encoding process. Affect.

  Therefore, although this image encoding device cannot be said to have a possibility of deteriorating the compression rate or the image quality because of the suppression of the calculation amount, the number of moving images to be encoded without particularly increasing the device cost. Any number of images can be encoded at the same time, and there is a useful effect that encoding of all moving images can be completed within a certain amount of time. In addition, as a method for preventing the deterioration of the image quality, a method of reducing the quantization step size related to the quantization process in the encoding process can be employed.

  In addition, the motion prediction processing method specifying means associates each numerical value and each parameter for specifying the processing method of the motion prediction processing with a relationship in which the amount of calculation of the motion prediction processing decreases as the number increases. It has a storage unit that pre-stores a table, and transmits a parameter in the table corresponding to the number acquired by the number acquisition unit to the motion prediction unit, so that the processing method of the motion prediction process is In particular, the motion prediction unit may perform a motion prediction process using the specified processing method based on the parameter transmitted by the motion prediction processing method specifying unit.

  As a result, the processing method of the motion prediction process can be specified using a table that has been appropriately adjusted in advance by experiments or the like. It is possible to change the processing method of the motion prediction process according to the number of moving images to be converted. It should be noted that the table is specified so that the total amount of computation required for motion prediction processing for each moving image is within a certain range regardless of the number of moving images to be encoded. If adjusted, it is possible to appropriately encode each moving image in parallel, regardless of whether the number of moving images to be encoded is two or three.

  In addition, the motion prediction processing method specifying unit determines a processing method related to the upper limit value of the number of reference image frames in the motion prediction process as the number of reference image frames increases according to the number acquired by the number acquisition unit. The motion prediction unit is specified by the motion prediction processing method specifying means for each processing target block in each image frame to be motion prediction processed in the motion prediction processing process. The motion vector may be searched from a number of reference image frames that do not exceed the upper limit value of the number of reference image frames in the processing method.

  Note that the moving image is composed of a plurality of image frames, and the motion prediction process searches for a reference image block having a high similarity in the reference image frame with respect to each processing target block divided within each image frame. This is a process for obtaining a motion vector indicating the position of the reference image block and obtaining difference data that is a difference between the image data of the processing target block and the reference image block, and the obtained data is an encoding process subsequent to the motion prediction process. Used in the process.

  Therefore, when the number of moving images to be encoded is large, the image encoding device according to the present invention limits the number of reference image frames in which the reference block is located in the motion prediction process for each moving image to a smaller extent. This reduces the amount of computation required to search for a reference block having a high degree of similarity with each processing target block in the image frame. As a result, the finite computation processing capability of the image coding apparatus is effectively utilized, and an arbitrary number of blocks can be obtained. The moving image can be encoded at the same time.

  Further, the motion prediction processing method specifying means narrows the search range as the number of processing methods related to the motion vector search range in the motion prediction processing increases according to the number acquired by the number acquisition means. The motion prediction unit specifies the search range in the processing method specified by the motion prediction processing method specifying means in the reference image frame for each processing target block in each image frame that is a motion prediction processing target. A motion vector pointing to the reference image block may be determined by searching for a reference image block similar to the processing target block.

As a result, if the number of moving images to be encoded is large, the search range of the reference block is narrowed in the motion prediction process for each moving image, so that the amount of calculation required for searching the reference block can be reduced. Can be encoded at the same time.
In addition, the motion prediction processing method specifying means determines the number of shape patterns as the number of processing methods related to various shape patterns of the reference image block in the motion prediction processing increases according to the number acquired by the number acquisition means. The motion prediction unit uses various shape patterns in the processing method specified by the motion prediction processing method specifying unit for each processing target block in each image frame that is a motion prediction processing target. Thus, a reference block in the reference image frame may be searched for a reference image block similar to the processing target block, and a motion vector indicating the reference image block may be determined.

As a result, if the number of moving images to be encoded is large, the number of matching shape patterns used when searching for reference blocks is reduced in the motion prediction process for each moving image. As a result, an arbitrary number of moving images can be encoded simultaneously.
The moving image acquisition unit acquires a new moving image even during the encoding process by the encoding unit, and the number acquisition unit determines the number of moving images to be encoded during the encoding process by the encoding unit. The motion prediction processing method specifying means may specify the motion prediction processing method each time the number of moving images is acquired by the number acquisition means.

As a result, even if the number of moving images input to the image encoding device changes, the motion prediction processing method is newly specified and the motion prediction processing is performed using that processing method accordingly. Can be encoded appropriately.
In addition, the moving image acquisition unit acquires a plurality of moving images to be encoded, and the image encoding device further receives receiving capability information indicating an arithmetic processing capability from an external device capable of executing the encoding process. And determining means for determining the number of moving images to be subjected to the encoding process requested to be executed from the external device based on the capability information; and determining by the determining means among the moving images acquired by the moving image acquiring means Transmitting means for transmitting the number of moving images to the external device, wherein the motion prediction processing method specifying means is responsive to a difference between the number acquired by the number acquiring means and the number determined by the determining means. Thus, the motion prediction processing method may be specified so that the amount of calculation decreases as the difference increases.

In addition, the determination unit determines a processing method of motion prediction processing in the encoding processing that requests execution to the external device, together with the number of moving images that are targets of the encoding processing that requests execution based on the capability information. The transmitting unit may further transmit information indicating the processing method of the motion prediction process determined by the determining unit to the external device.
As a result, encoding of a plurality of moving images can be appropriately distributed and processed by a plurality of devices.

  Further, the image encoding device further specifies a quantization step size according to the number acquired by the number acquisition means so that the quantization step size decreases as the number increases. The encoding unit may include a quantization unit that quantizes the difference data obtained as a result of the motion prediction process with the quantization step size specified by the quantization step size specifying unit. Good.

As a result, as the number of moving images to be encoded increases, image quality degradation that can occur by suppressing the amount of motion prediction processing for each moving image is reduced, and the quantization step size related to the quantization processing in the encoding processing is reduced. Interpolation is possible by improving the image quality.
An image encoding apparatus according to the present invention is an image encoding apparatus that performs an encoding process to compress a moving image, a number acquisition unit that acquires the number of moving images to be encoded, and an encoding target A video acquisition unit that acquires one or a plurality of videos and a processing method of an encoding process that affects the calculation amount of the encoding process are calculated as the number increases according to the number acquired by the number acquisition unit. The processing method specifying means for specifying the amount to be reduced and the processing method specified by the processing method specifying means perform the encoding process on the moving image acquired by the moving image acquisition means, and the acquired moving image If it is plural, it is provided with the encoding means which performs an encoding process by time division for each moving image.

According to the image encoding device according to the present invention, the amount of calculation required for encoding processing for each moving image is adjusted according to the number of moving images to be encoded in parallel. Any number of moving images can be encoded at the same time.
Further, the processing method specifying means includes a table in which each numerical value and each parameter for specifying the processing method of the encoding process are associated with a relationship in which the calculation amount of the encoding process decreases as the number increases. It has a storage unit that stores in advance, and specifies the processing method of the encoding process by transmitting the parameters in the table corresponding to the number acquired by the number acquisition unit to the encoding unit. The encoding unit may perform the encoding process by the specified processing method based on the parameter transmitted by the processing method specifying unit.

  As a result, the processing method of the encoding process can be specified using a table that has been appropriately adjusted in advance through experiments or the like. The processing method of the encoding process can be changed according to the number of moving images to be converted. Note that the table is specified so that the total amount of computation required for the encoding process for each moving image is within a certain range regardless of the number of moving images to be encoded. If adjusted, it is possible to appropriately encode each moving image in parallel, regardless of whether the number of moving images to be encoded is two or three.

  In addition, the encoding unit includes an intra encoding unit that performs an intra-image frame encoding process on a part or all of image frames in a moving image to be processed as a part of the encoding process, and the processing method The specifying means reduces the number of search directions as the number increases in accordance with the number acquired by the number acquisition means for the processing method related to the direction for searching for a predicted value in intra-frame coding. And the intra coding unit includes, for each processing target block in the image frame, the processing target block of each processing target block along each of the search directions for the predicted value in the processing method specified by the processing method specifying means. It is good also as searching for the predicted value most similar to image data.

  Here, the intra-frame encoding process searches for a predicted value that is most similar to the image data of the processing target block among the predicted values calculated based on blocks in a predetermined direction within the same image frame as the processing target block. This is an intra encoding process for obtaining difference data that is the difference between the predicted value of the search result and the image data of the processing target block, and the obtained data is used in an encoding process step subsequent to the intra encoding process.

As a result, if the number of videos to be encoded is large, in the intra encoding process for each video, the calculation target range for calculating the prediction value is reduced, so that the amount of calculation can be reduced. The moving image can be encoded at the same time.
In addition, the processing method specifying means determines whether the processing method relating to whether or not to use inter-frame coding is used between image frames if the number is less than a predetermined number according to the number acquired by the number acquisition means. If the number is greater than or equal to a predetermined number using encoding, it is specified not to use inter-frame image encoding, and the encoding means specifies that the processing method specified by the processing method specifying means is an image inter-frame encoding. If the processing method is to be used, inter-frame coding and intra-frame coding are performed on the moving image to be encoded, and the processing method specified by the processing method specifying means does not use inter-frame coding. In the case of a processing method, intra-frame encoding may be performed without performing inter-frame encoding on a moving image to be encoded.

As a result, only the intra-frame encoding is performed when the number is large, so that the amount of calculation can be reduced and an arbitrary number of moving images can be encoded simultaneously.
Further, the encoding means includes a variable length encoding unit that performs variable length encoding processing as a part of the encoding processing, and the processing method specifying means includes the processing method related to variable length encoding, In accordance with the number acquired by the number acquisition means, if the number is less than a predetermined number, the calculation coding is used, and if the number is a predetermined number or more, it is specified not to use arithmetic coding, and the variable length code If the processing method specified by the processing method specifying unit uses arithmetic coding, the conversion unit performs variable length coding by arithmetic coding, and the processing method specified by the processing method specifying unit is If the processing method does not use arithmetic coding, variable length coding may be performed based on a reference table related to a predetermined code word without performing arithmetic coding.

As a result, the amount of computation can be reduced because variable length coding is performed based on the table without using arithmetic coding, which is generally considered to have a large amount of computation when the number is large, and as a result, any number of videos can be encoded simultaneously. It becomes possible to become.
The moving image acquisition unit acquires a new moving image even during the encoding process by the encoding unit, and the number acquisition unit determines the number of moving images to be encoded during the encoding process by the encoding unit. The processing method specifying means may specify the processing method of the encoding process every time the number of moving images is acquired by the number obtaining means.

As a result, even if the number of moving images input to the image encoding device changes, the processing method of the encoding process is newly specified in accordance with the change, and the encoding process is performed according to that processing method. Can be encoded properly.
In addition, the moving image acquisition unit acquires a plurality of moving images to be encoded, and the image encoding device further receives receiving capability information indicating an arithmetic processing capability from an external device capable of executing the encoding process. And determining means for determining the number of moving images to be subjected to encoding processing to be requested to be executed by the external device based on the capability information, and determining by the determining means among the moving images acquired by the moving image acquiring means Transmission means for transmitting the number of videos to the external device, and the processing method specifying means according to a difference between the number acquired by the number acquisition means and the number determined by the determination means, The processing method of the encoding process may be specified so that the calculation amount decreases as the difference increases.

Thereby, it becomes possible to appropriately distribute and process encoding of a plurality of moving images with a plurality of devices.
The image encoding apparatus further includes image quality information acquisition means for acquiring image quality information indicating required image quality, and the processing method specifying means acquires the processing method of encoding processing by the number acquisition means. The number may be specified in accordance with the image quality information acquired by the image quality information acquisition unit.

For example, when the required image quality is high, the quantization step size is reduced and the request is met, so that encoding according to the required image quality is possible.
In addition, the image encoding device further includes a decoding unit that decodes a compressed moving image by sharing a part of the circuit with the encoding unit, and the processing method specifying unit is further decoded by the decoding unit. The processing method of the encoding process may be specified so that the amount of calculation decreases as the amount of moving images to be decoded increases according to the amount of moving images to be decoded.

As a result, any number of moving images can be encoded in parallel with the decoding of the moving images.
An image encoding method according to the present invention is an image encoding method for performing an encoding process including a motion prediction process in order to compress a moving image, and the number acquiring step for acquiring the number of encoding target moving images. And a moving image acquisition step for acquiring one or a plurality of moving images to be encoded, and a motion prediction processing method that affects the calculation amount of the motion prediction processing, according to the number acquired by the number acquisition step, The motion prediction processing method specifying step that specifies the amount of calculation to decrease as the number increases, and the encoding processing is performed on the moving image acquired by the moving image acquisition step, and if there are a plurality of acquired moving images, each An encoding step that performs time-division encoding processing on a moving image, and the encoding step includes the motion prediction processing method specifying step as a part of the encoding processing. Characterized in that it comprises a motion estimation sub-step of performing motion prediction process at the specified processing method by.

  An image encoding method according to the present invention is an image encoding method for performing an encoding process to compress a moving image, the number acquiring step for acquiring the number of moving images to be encoded, and an encoding target The moving image acquisition step for acquiring one or a plurality of moving images and the processing method of the encoding process that affects the calculation amount of the encoding process are calculated as the number increases according to the number acquired by the number acquisition step. The processing method specifying step for specifying the amount to be reduced, and the processing method specified by the processing method specifying step, performing the encoding process on the moving image acquired by the moving image acquisition step, and the acquired moving image If it is plural, it includes an encoding step of performing encoding processing in a time division manner for each moving image.

  According to these image encoding methods, the amount of calculation required for the encoding process is adjusted according to the number of moving images to be encoded in parallel, so that an arbitrary number of moving images can be processed with a certain amount of calculation amount within a certain range. Can be encoded.

It is a block diagram of the image coding apparatus 100 which concerns on Embodiment 1 of this invention. It is a figure which shows the structure and example of a content of a motion prediction processing system table. It is a figure which shows the structure and example of a content of an intra-encoding process system table. 4 is a flowchart showing processing method specifying processing by a processing method specifying unit 130; 5 is a flowchart showing an encoding process by an encoding unit 140. 1 is a configuration diagram of an image encoding device 200. FIG. It is a block diagram of the image coding system which concerns on Embodiment 2 of this invention. 5 is a flowchart illustrating request processing performed by the image encoding device 300.

Explanation of symbols

100, 200, 300, 400 Image encoding device 110 Number acquisition unit 120 Movie acquisition unit 130, 230 Processing method specifying unit 131 Motion prediction processing method specifying unit 132 Intra coding processing method specifying unit 133, 233 Quantization step size specifying unit 134 Variable Length Coding Processing Method Identification Unit 140 Encoding Unit 141 Video Selection Unit 142 Motion Prediction Unit 143 Intra Coding Unit 144 Frequency Conversion Unit 145 Quantization Unit 146 Variable Length Coding Unit 147 Output Unit 150 Recording Medium 211 Image Quality Acquisition Unit 301, 402 Processing amount management unit

<Embodiment 1>
Hereinafter, the image coding apparatus 100 according to the first embodiment of the present invention will be described.
<Configuration>
FIG. 1 is a configuration diagram of an image encoding device 100 according to Embodiment 1 of the present invention.
As illustrated in FIG. 1, the image encoding device 100 includes a number acquisition unit 110, a moving image acquisition unit 120, a processing method identification unit 130, and an encoding unit 140.

In the figure, there is also shown a recording medium 150 such as an optical disk for accumulating data output by the image encoding apparatus 100 encoding the moving image. Here, the moving image is also used as a term indicating moving image data including an image frame representing the image content displayed at the time of reproduction.
Here, the number acquisition unit 110 is an input interface that receives an input of the number of moving images to be encoded from the outside of the image encoding device 100 and transmits the input to the processing method specifying unit 130.

  The moving image acquisition unit 120 is a moving image input interface, and has a plurality of buffer memories such as a first buffer, a second buffer,..., An Nth buffer having a capacity capable of storing a predetermined number of image frames constituting the moving image. It has a function of receiving input of one or a plurality of moving images to be encoded from the outside of the image encoding device 100 and storing each moving image in each internal buffer memory. Since the moving image is so-called stream data, when the total data amount of the moving image is large, the moving image acquisition unit 120 continues even while the image encoding device 100 is partially encoding the moving image. Is acquired.

  The processing method specifying unit 130 is realized by a memory and a processor, and includes, as functional components, a motion prediction processing method specifying unit 131, an intra coding processing method specifying unit 132, a quantization step size specifying unit 133, and a variable length coding process. A function that has a method specifying unit 134, specifies a processing method of the encoding process in the encoding unit 140 according to the number transmitted from the number acquisition unit 110, and transmits a parameter indicating the processing method to the encoding unit 140 Have

  The encoding unit 140 is realized by a memory, each circuit group that handles each part of the encoding process, and a control processor that controls each circuit group, and includes a moving image selection unit 141, a motion prediction unit 142, and an intra encoding unit 143. , A frequency conversion unit 144, a quantization unit 145, a variable length encoding unit 146, and an output unit 147, and has a function of reading a video from the buffer memory of the video acquisition unit 120, encoding it, and recording it on the recording medium 150. Here, this encoding is described as being performed in conformity with the MPEG4 AVC standard.

The motion prediction processing method specifying unit 131 of the processing method specifying unit 130 has a motion prediction processing method table that is one area of the memory, and based on the number transmitted from the number acquisition unit 110 based on the motion prediction processing method table. Thus, the motion prediction unit 142 has a function of specifying a processing method of motion prediction processing. The motion prediction method table will be described in detail later.
The intra coding processing method specifying unit 132 has an intra coding processing method table that is an area of the memory, and based on the intra coding processing method table, according to the number transmitted from the number obtaining unit 110, the intra coding processing method table is used. The encoding unit 143 has a function of specifying a processing method for intra encoding processing. The intra coding method table will be described in detail later.

  The quantization step size specifying unit 133 has a function of calculating a quantization step size used for quantization in the quantization unit 145 by performing a predetermined calculation based on the number transmitted from the number acquisition unit 110. . This predetermined operation is a predetermined operation so that the quantization step size decreases as the number increases, and when the number is 1, it is equivalent to a conventional practical image encoding device. This is an operation determined to have a quantization step size of about.

The variable-length encoding processing method specifying unit 134 has a function of specifying a variable-length encoding processing method based on the number transmitted from the number acquisition unit 110.
Specifically, this function uses a method of encoding based on a pre-established event / codeword comparison table assuming the statistics of events to be variable-length encoded, or further reduces the compression rate. This is a function for specifying whether to use a method called arithmetic coding in which encoding is performed while dynamically determining a code word by arithmetic operation based on a past event occurrence probability when encoding is performed in order to enhance.

  In addition, the moving image selection unit 141 of the encoding unit 140 acquires the number of moving images to be encoded via the processing method specifying unit 130, and according to the number of the moving image selection unit 141, from the buffer memory for the number of the moving image acquisition unit 120 A moving image is read cyclically every time, and each moving image is assumed to be a series of digitized image frames, and an image frame group constituting each moving image is transmitted to the motion prediction unit 142 or the intra coding unit 143. This is a circuit having a function of starting the encoding process in a time division manner for each moving image.

In this time division processing, for example, when there are two moving images, the moving image selection unit 141 switches the moving images to be read from the buffer memory alternately for each image frame, and sequentially selects a predetermined size such as a so-called macroblock unit. This is realized by transmitting to the motion prediction unit 142 or the intra coding unit 143 in block units.
The motion prediction unit 142 can access the buffer memory of the video acquisition unit 120 for each video, receives a parameter indicating the processing method of the specified motion prediction process from the motion prediction processing method specifying unit 131, and the processing method Thus, the circuit has a function of executing a motion prediction process in units of blocks of a predetermined size with respect to a moving image frame selected by the moving image selection unit 141.

  The motion prediction process is a so-called inter-frame image encoding process. For each block in the image frame, the block (hereinafter referred to as “processing target block”) in another image frame having a high degree of similarity. A block (hereinafter referred to as a “reference block”) is searched, a motion vector indicating the position of the reference block is calculated, and difference data that is a difference value of image data between the processing target block and the reference block of the search result is calculated. This is a process of calculating and transmitting difference data to the frequency converter 144. Here, the high similarity between the image data of the processing target block and the reference block means that the distribution tendency of the image data of both blocks is similar, and the similarity is the highest when the image data of both blocks match. high.

The intra coding unit 143 receives a parameter indicating the processing method of the specified intra coding process from the intra coding processing method specifying unit 132, and the image frame of the moving image selected by the moving image selection unit 141 according to the processing method Is a circuit having a function of executing intra coding processing in units of blocks of a predetermined size in the image frame.
Intra coding processing is so-called image frame coding processing. For each block in the image frame, the image data of the block and blocks in several predetermined directions around the same image frame This is a process of searching for a predicted value that is most similar to a predicted value based on, calculating difference data that is a difference value between the image data and the predicted value, and transmitting the difference data to the frequency converter 144. .

The frequency transform unit 144 performs orthogonal transform on the difference data transmitted from the motion prediction unit 142 or the intra coding unit 143 so as to suppress the numerical value constituting the difference data to be small by removing the spatial redundancy. , A circuit having a function of transmitting the result of the orthogonal transformation to the quantization unit 145.
The quantization unit 145 obtains the difference data after being orthogonally transformed by the frequency conversion unit 144, and in order to suppress the numerical values constituting the difference data to be smaller, the quantization step size specifying unit 133 4 is a circuit having a function of performing quantization by rounding the result obtained by dividing by the quantization step size specified in (1) to an integer value and transmitting the quantization result to the variable-length encoding unit 146.

The variable length coding unit 146 obtains the data obtained as a result of the quantization performed by the quantization unit 145, and obtains the data amount to further compress the data amount by expressing the information with higher appearance frequency with a shorter code. This is a circuit having a function of performing variable-length coding on data by the processing method specified by the variable-length encoding processing method specifying unit 134 and transmitting the result to the output unit 147.
The output unit 147 is a circuit having a function of recording the encoded data transmitted from the variable length encoding unit 146 on the recording medium 150 while distinguishing the data for each moving image.

<Data>
Hereinafter, the table data stored in the processing method specifying unit 130 will be described.
FIG. 2 is a diagram illustrating a configuration and example contents of the motion prediction processing method table.
As shown in the figure, the motion prediction processing method table included in the motion prediction processing method specifying unit 131 associates the number of moving images 51 with the reference image frame number upper limit 52, the reference image block search range 53, the search pattern 54, and the like. It is a table. Note that the number 51 of moving images takes a value from 1 to a predetermined N, for example, four.

Here, the reference image frame number upper limit 52 indicates the maximum number allowed as the number of image frames referred to by the processing target block in a certain image frame in the motion prediction process.
The reference image block search range 53 indicates the search range to which to search based on the position of the processing target block based on the range in which a reference image block having a high similarity with a certain processing target block is searched.

  The search pattern 54 indicates the size and shape of the processing target block and the reference image block. Specifically, the search pattern 54 is a pattern of 16 pixels wide × 16 pixels high, a pattern of 16 pixels wide × 8 pixels high, Pattern of width 8 pixels x height 16 pixels, pattern of width 8 pixels x height 8 pixels, pattern of width 8 pixels x height 4 pixels, pattern of width 4 pixels x height 8 pixels, width 4 pixels x height This indicates which of the 7 patterns of 4 pixels is used to search for the reference image block.

  According to the motion prediction processing method table illustrated in FIG. 2, when there is one moving image, the upper limit is two reference image frames, and the search range of the reference image block is from −15 pixels based on the processing target block. It is possible to specify a motion prediction processing method in which the range is +15 pixels, and all seven patterns can be applied as a set of search pattern size and shape. When there are two moving images, the upper limit is one reference image frame, the search range of the reference image block is a range from −15 pixels to +15 pixels based on the processing target block, and the size and shape of the search pattern 4 patterns of a pattern of width 16 pixels × height 16 pixels, a pattern of width 16 pixels × height 8 pixels, a pattern of width 8 pixels × height 16 pixels, and a pattern of width 8 pixels × height 8 pixels. The processing method of the motion prediction process that can be applied can be specified.

FIG. 3 is a diagram illustrating a configuration and an example of contents of an intra coding processing method table.
The intra coding processing method table included in the intra coding processing method specifying unit 132 is a table in which the number of moving images 61 and the prediction direction 62 are associated with each other as illustrated in FIG. Note that the number 61 of moving images takes a value from 1 to N, for example, a predetermined number.
Here, the prediction direction 62 indicates the upper limit of the direction in which a prediction value is obtained for a block in a certain image frame in the intra coding process. This direction is any one of nine directions from prediction mode 0 to prediction mode 8 for a block of 4 pixels wide × 4 pixels high.

  That is, with respect to the prediction target block, with reference to the block, prediction mode 0 for determining the value of the upper block as a prediction value, prediction mode 1 for determining the value of the left block as a prediction value, and the value of the upper block Predictive mode 2 in which the average value of the block values in the left direction and the value in the left direction is determined as the prediction value, and the prediction value is set in the 45 degree direction using the block value in the upper direction and the block value in the upper right direction for the prediction target block. Prediction mode 3 to be determined, Prediction mode 4 to determine the prediction value in the 135 degree direction using the value of the block in the upward direction and the value of the block in the left direction for the prediction target block, and similarly, the prediction value is set in the direction of about 112 degrees Prediction mode 5 to be determined, similarly prediction mode 6 to set a prediction value in the direction of about 158 degrees, prediction mode 7 to set the prediction value in the direction of about 68 degrees, prediction to set the prediction value in the direction of about 203 degrees Prediction direction 62 of the modes 8 is selected.

According to the intra coding processing method table illustrated in FIG. 3, nine directions of prediction modes 0 to 8 are applicable when there is one moving image, and prediction modes 0 to 8 are applicable when there are two moving images. The processing method of the intra-encoding processing that can apply the five directions of 4 can be specified.
<Operation>
Hereinafter, the operation of the image coding apparatus 100 having the above-described configuration will be described.

The image encoding device 100 receives an input of an encoding target moving image and the number of moving images from the outside via an input interface, and the processing method specifying unit 130 specifies a processing method of encoding processing based on the number of moving images. A processing method specifying process is executed, and an encoding process for each moving image is executed by the encoding unit 140 according to the result.
FIG. 4 is a flowchart showing the processing method specifying process by the processing method specifying unit 130. Hereinafter, the operation of the processing method specifying unit 130 will be described with reference to FIG.

  The processing method specifying unit 130 first acquires the number of encoding target moving images from the number acquisition unit 110 (step S11), and the motion prediction processing method specifying unit 131 performs motion according to the number based on the motion prediction processing method table. Each parameter indicating the prediction processing method is specified (step S12), and the parameter indicating the intra encoding processing method is specified according to the number by the intra encoding processing method specifying unit 132 based on the intra encoding processing method table ( Step S13).

  Further, the processing method specifying unit 130 calculates a quantization step size by performing a predetermined calculation according to the number by the quantization step size specifying unit 133 (step S14), and the variable length coding processing method specifying unit 134. By comparing the number and a predetermined threshold value set in advance as 2 for example (step S15), if the number is less than the threshold value, a variable length coding processing method is shown so that arithmetic coding is used. A parameter is specified (step S16), and if the number is equal to or greater than a threshold, a parameter indicating a variable length coding processing method is specified so that arithmetic coding is not used (step S17). Parameters indicating the processing method, parameters indicating the intra coding processing method, quantization step size, and parameters indicating the variable length coding processing method. The meter is transmitted to the encoding unit 140 (step S18), and ends the processing method specifying process.

By such processing method specifying processing, the processing method is specified so that the larger the number of moving images to be encoded, the smaller the processing amount of the encoding processing.
However, the quantization step size related to quantization becomes smaller as the number of moving images to be encoded increases, and as a result, the amount of quantization processing increases. Note that the amount of quantization processing, for example, the amount of computation when executed by the processor is sufficiently small in the practical range compared to the motion prediction processing, and the number of processing increases due to the processing method specifying processing described above. In this case, the amount of motion prediction processing is reduced, so that the image quality tends to deteriorate. In this case, in order to compensate for the deteriorated image quality, the quantization step size is reduced in order to reduce the image quality deterioration due to quantization. Like to do.

FIG. 5 is a flowchart showing the encoding process by the encoding unit 140. Hereinafter, the operation of the encoding unit 140 will be described with reference to FIG.
First, according to the number transmitted from the processing method specifying unit 130, the moving image selection unit 141 of the encoding unit 140 has the number of moving images sequentially acquired from the outside by the moving image acquisition unit 120 and temporarily accumulated. From each buffer memory, moving images are switched and read sequentially for each image frame (step S21).

  Subsequently, the encoding unit 140 can determine whether the image frame of the moving image read by the moving image selection unit 141 is an image frame targeted only for the intra coding process or a target of the motion prediction process according to a predetermined criterion. It is determined whether it is an image frame (step S22), and if it is an image frame that is only subject to intra coding processing, the image is within the prediction direction according to the parameters specified by the intra coding processing method specifying unit 132. The intra encoding unit 143 performs an intra encoding process by searching for an optimal prediction value for each block in the frame and obtaining difference data that is a difference between the image data of the block and the prediction value (step S23).

  If it is determined in step S22 that the image frame is not the target of only the intra coding process, the processing target block is equal to or less than the upper limit number indicated by the parameter according to the parameter specified by the motion prediction processing method specifying unit 131. The reference block is searched for from the search range of the reference image block in the reference image frame indicated by the parameter, and the search pattern used for searching the reference block among the processing target blocks is the parameter. The reference image block having the shape and size shown in FIG. 6 is searched for a reference image block having a high degree of similarity to the processing target block, and a motion vector indicating the position of the reference image block is obtained. The reference image block and the processing target block To obtain difference data, which is the difference between image data and The measuring process performed by the motion prediction unit 142 (step S24).

Note that when the reference image block is referred to from a plurality of image frames because the parameter indicating the upper limit of the number of reference image frames is 2 or more, the average value of the image data for the reference image block searched from each image frame is The difference from the image data of the processing target block is obtained as difference data.
Further, following step S23 or step S24, the frequency conversion unit 144 of the encoding unit 140 performs orthogonal transform on the obtained difference data (step S25), and the quantization unit is performed on the resultant data. 145 performs quantization with the quantization step size specified by the quantization step size specifying unit 133 (step S26).

  Following the quantization, the variable length encoding unit 146 uses the variable length code for the quantized data according to the parameters specified by the variable length encoding processing method specifying unit 134 with or without arithmetic coding. (Step S27), and the output unit 147 records the data after variable length coding on the recording medium 150 by distinguishing it for each moving image (step S28), thereby encoding the image frame (step S22). ˜S28) ends.

  When there are a plurality of moving images, that is, when data indicating a plurality of numbers is acquired by the number acquisition unit 110, after an encoding process is performed on a certain image frame in the first moving image, After the encoding process (steps S22 to S28) is performed, the encoding process is sequentially performed for the number of moving images, and the encoding process for the next image frame is started for each moving image from the first moving image. The encoding process is performed cyclically.

When data indicating one is acquired by the number acquisition unit 110, encoding processing is sequentially performed on each image frame for one moving image. Each process (steps S22 to S28) in the encoding process illustrated in FIG. 5 is realized by pipeline processing in units of blocks.
<Modification>
Hereinafter, an image encoding device 200 that is a modification of the image encoding device 100 will be described.

The image coding apparatus 200 is an apparatus to which the image coding apparatus 100 is added with a function of accepting an input of a requested image quality from the outside and coding while maintaining the image quality corresponding to the request.
FIG. 6 is a configuration diagram of the image encoding device 200.
The image encoding device 200 includes an image quality acquisition unit 211, a number acquisition unit 110, a moving image acquisition unit 120, a processing method identification unit 230, and an encoding unit 140, as shown in FIG. Of the constituent elements of the image encoding apparatus 200, the same constituent elements as those of the image encoding apparatus 100 are denoted by the same reference numerals in FIG. 6 as those in FIG. 1, and detailed descriptions thereof are omitted here. To do.

Here, the image quality acquisition unit 211 is an input interface having a function of acquiring image quality information indicating image quality input by a user or the like and transmitting the acquired image quality information to the processing method specifying unit 230. The image quality information is, for example, numerical information that expresses the height of the image quality by the size of a number.
Further, the processing method specifying unit 230 includes a motion prediction processing method specifying unit 131, an intra coding processing method specifying unit 132, a quantization step size specifying unit 233, and a variable length coding processing method specifying unit 134.

  The quantization step size specifying unit 233 performs a predetermined calculation based on the number transmitted from the number acquisition unit 110 and the image quality information transmitted from the image quality acquisition unit 211, thereby performing the quantization in the quantization unit 145. It has a function of calculating the quantization step size used for the conversion. This predetermined calculation is a predetermined calculation so that the quantization step size decreases as the number increases, and the quantization step size decreases as the image quality indicated by the image quality information increases. .

Therefore, according to the image encoding device 200, the quantization step size is specified reflecting the image quality information acquired by the image quality acquisition unit 211, and the quantization unit 145 performs quantization with the quantization step size. Therefore, encoding is performed while maintaining the requested image quality.
<Embodiment 2>
Hereinafter, an image coding system according to Embodiment 2 of the present invention will be described.

<Configuration>
FIG. 7 is a configuration diagram of an image coding system according to Embodiment 2 of the present invention.
This image encoding system is a system in which encoding processing for a moving image is distributed by a plurality of image encoding devices, and as shown in the figure, by an image encoding device 300 and an image encoding device 400 connected to a network. Composed. Among the components of these image encoding devices, the same components as those of the image encoding device 100 shown in the first embodiment are denoted by the same reference numerals as those in FIG. 1 in FIG. Detailed description is omitted here.

The image encoding device 300 obtains information indicating the number of moving images and the moving image from the outside, and causes the image encoding device 400 to share the encoding processing for some moving images while also encoding the other moving images. The number acquisition unit 110, the moving image acquisition unit 120, the processing method identification unit 130, the encoding unit 140, and the processing amount management unit 301 are provided.
In addition, the image encoding device 400 includes a processing method specifying unit 130, an encoding unit 140, and a processing amount management unit 402.

  The processing amount management unit 301 of the image encoding device 300 has a memory in which processing amount information indicating the processing capability of the device itself is stored in advance, and inquires the processing capability of the image encoding device 400 to obtain a response. Based on the processing capability, information indicating the number acquired from the outside by the number acquisition unit 110, and the processing capability of the own device, the number of moving image encoding processing requests to the image encoding device 400. The information indicating the number of lines determined to be determined and requested and the moving images acquired from the outside by the moving image acquiring unit 120 for the number are transmitted to the image encoding device 400, and the remaining moving images are image-coded. The encoding apparatus 300 has a function of transmitting the remaining number to the processing method specifying unit 130 in the image encoding apparatus 300.

  The processing amount management unit 402 of the image encoding device 400 has a memory in which processing amount information indicating the processing capability of the own device is stored in advance, and when a processing capability inquiry is received from the image encoding device 300, When receiving the processing amount information of its own device with reference to the memory and the information indicating the number from the image encoding device 300, the number of the information is stored in the image encoding device 400. It has a function of transmitting to the processing method specifying unit 130.

The processing amount information indicating the processing capacity stored in the image encoding device 300 and the image encoding device 400 described above is, for example, a processing method of a predetermined encoding process used when encoding one moving image. When encoding is performed with uniform application, the number of moving images can be encoded at the same time.
<Operation>
Hereinafter, the operation of the image coding system having the above-described configuration will be described.

When a plurality of moving images and information indicating the number of the moving images are input from the outside to the image encoding device 300, the image encoding device 300 requests the image encoding device 400 to encode some moving images. Then, a request process for encoding the remaining moving image is executed by the own device.
FIG. 8 is a flowchart showing request processing performed by the image encoding device 300. Hereinafter, the operation of the image encoding device 300 and the operation of the image encoding device 400 in response thereto will be described with reference to FIG.

The image encoding device 300 inquires about the processing capability of the image encoding device 400 connected to the network, and receives processing amount information transmitted by the image encoding device 400 in response to this (step S31).
Subsequently, the processing amount management unit 301 of the image encoding device 300 obtains a ratio of the processing capability value indicated by the received processing amount information to the processing capability value indicated by the processing amount information of the own device, and calculates the ratio as the number. An integer closest to the value obtained by multiplying the number acquired by the acquisition unit 110 is determined as the requested number (step S32), and the requested number is transmitted to the image encoding device 400 to request encoding ( Step S33).

  As a result, the requested number is received by the image encoding device 400, and the processing method specifying unit 130 in the image encoding device 400 specifies each processing method of the encoding process according to the requested number. In addition, the encoding unit 140 in the image encoding device 400 receives the requested number of moving images from the image encoding device 300, and specifies each received moving image by the processing method specifying unit 130 in a time-sharing manner. The encoding process is executed in accordance with the processed processing method.

Subsequent to step S <b> 33, the processing amount management unit 301 of the image encoding device 300 determines the number that is the difference between the number acquired by the number acquisition unit 110 and the requested number as a processing method specifying unit in the image encoding device 300. 130, the request process is terminated (step S34). As a result of step S34, the processing method specifying unit 130 of the image encoding device 300 specifies each processing method of the encoding process according to the number of differences. Then, according to each processing method, the encoding unit 140 of the image encoding device 300 performs encoding processing for the number of moving images corresponding to the difference.
<Supplement>
The image encoding device according to the present invention has been described above based on Embodiments 1 and 2. However, the image encoding device can be modified as follows, and the present invention is limited to the image encoding device described in the above embodiment. Of course not.
(1) Although the image encoding apparatus shown in Embodiments 1 and 2 records the encoded moving image on the recording medium 150, the recording medium 150 includes a CD-ROM (Compact Disk Read Only Memory), The optical disk is not limited to an optical disk such as a DVD (Digital Versatile Disk) or BD (Blu-ray Disk), but may be a hard disk, a memory card, or the like. Further, the output unit 147 of the image encoding device may send each encoded moving image to the network instead of recording on the recording medium.
(2) Although the image encoding apparatus shown in the first and second embodiments performs encoding in conformity with the MPEG4 AVC standard, for example, even if it conforms to the MPEG2 video standard, MPEG4 visual standard, or the like. Well, the scope of application of the present invention is not particularly limited to a specific standard.
(3) The processing method specifying unit of the image encoding device described in the first and second embodiments includes a motion prediction processing method, an intra encoding processing method, a quantization processing method method according to a quantization step size, and The parameters that specify each of the processing methods of the variable-length encoding process are specified, but in addition to these, for example, if the number of moving images to be encoded is 3 or more, a predetermined number or more It is also possible to specify whether or not to execute the motion prediction process so that only the intra encoding process is performed without performing the motion prediction process, and the processing method of the encoding process not exemplified in the first and second embodiments May be changed according to the number of moving images to be encoded.

Conversely, not all of the motion prediction processing method, intra encoding processing method, and variable length encoding processing method, but only a part of the processing methods are specified, and the video to be encoded is specified. The processing method specifying unit and the encoding unit may be configured so that only the encoding processing part related to a part of the processing methods varies depending on the number of the processing methods.
(4) The processing method specifying unit in the image encoding device described in the first and second embodiments includes a motion prediction method table and an intra encoding processing method table, and motion prediction in the encoding process using these tables. Each processing method for processing and intra coding processing is specified. However, a lookup table is similarly used to specify the processing method for quantization processing related to the quantization step size and the processing method for variable length coding processing. The parameters indicating each processing method may be specified by calculation based on a predetermined calculation formula.

However, as the number of moving images to be encoded increases, the processing method is specified so that the amount of motion prediction processing, intra coding processing, and variable length coding processing for one moving image tends to decrease. The quantization step size needs to be specified to be smaller as the number of moving images is increased in order to suppress degradation in image quality to some extent by reducing this processing amount. .
(5) Although the quantization step size specifying unit 233 of the image encoding device shown in the modification of the first embodiment specifies the quantization step size based on the image quality information indicating the required image quality and the number, The quantization step size may be specified based on only the image quality information without depending on the number.
(6) The specific content of the processing method specifying unit in the image coding apparatus shown in FIGS. 2 to 4 in Embodiments 1 and 2 specifying the processing method according to the number is merely an example. It can be modified depending on the circuit configuration and processing performance of the encoding unit.

In order to make it practical, processing is performed so that the processing speed of the encoding process executed by each circuit of the encoding unit is within a certain range regardless of the number of moving images to be encoded. In order to specify the method, it is desirable that a table or an arithmetic expression used for specifying the processing method in the processing method specifying unit is determined based on experiments or theoretical calculations.
(7) Each function unit constituting the encoding unit in the image encoding apparatus shown in the first and second embodiments is not limited to a circuit including only hardware, and the function is realized by executing software including software. In this case, the processor that executes the software may be provided for each functional unit, or a function corresponding to a plurality of functional units may be realized by one processor.

The image encoding apparatus may be realized as part or all of LSI (Large Scale Integration), VLSI (Very Large Scale Integration), etc., or may be realized by a plurality of LSIs, etc. You may implement | achieve by LSI etc. and other circuits.
(8) The number acquisition unit 110 of the image encoding device shown in the first embodiment acquires the number at any time even after the encoding unit 140 has started encoding processing for a moving image, The method specifying unit 130 may newly specify a processing method when the number changes, and the encoding unit 140 may perform encoding according to the result.

As a result, it is possible to cope with a case where the number of input moving images varies.
(9) The image coding apparatus shown in the first embodiment is modified, and a circuit that can be used for decoding in the coding unit 140 is also used for decoding, and further, a circuit necessary for decoding is further used. In addition, a decoding function for decoding the compressed moving image may be provided.

In this case, the processing method specifying unit 130 may be modified to specify the processing method so that the amount of calculation related to encoding for each moving image decreases as the amount of moving images to be decoded increases.
(10) The image encoding device 300 shown in the second embodiment may be modified to include a functional unit corresponding to the processing amount management unit 402 of the image encoding device in addition to the original configuration requirements. According to the modified image encoding device, the image encoding device 400 itself can request the image encoding device 400 to perform the encoding process, and when the other image encoding device 300 requests the encoding process, The moving image can be acquired and encoded through the moving image acquisition unit.
(11) The image coding system shown in the second embodiment is modified to exclude the processing method specifying unit 130 from the image encoding device 400, and the processing method specifying unit 130 of the image encoding device 300 is acquired by the number acquisition unit. The processing amount management unit 301 determines the number of requests to the image encoding device 400 and then determines the requested number of requests and the processing method. The processing method specified by the specifying unit 130 is transmitted to the image encoding device 400, and the encoding unit 140 of the image encoding device 400 generates the requested number of moving images according to the processing method received from the image encoding device 300. Encoding may be performed.
(12) A program for causing a processor to execute each process (see FIGS. 4 and 5) in the image encoding device can be recorded on a recording medium or distributed and distributed via various communication paths. Such recording media include IC cards, hard disks, optical disks, flexible disks, ROMs, and the like. The distributed and distributed program is used by being stored in a memory or the like that can be read by the processor, and the processor executes the program, thereby realizing the function of the image encoding device described in the first embodiment. Will come to be.

  The image encoding apparatus according to the present invention can be mounted and used in a hard disk recorder, a DVD recorder, or the like.

[Document Name] Description [Title of Invention] Image Encoding Device [Technical Field]
[0001]
The present invention relates to a moving image compression coding technique, and more particularly, to an image coding device that compresses and codes a plurality of moving images.
[Background]
[0002]
Conventionally, MPEG (Moving Picture Experts Group) standard, for example, MPEG2 video standard (ISO) is used as a standard for compression encoding (hereinafter simply referred to as “encoding”) used when transmitting moving images or recording on a recording medium. / IEC18818-2), MPEG4 visual standard (ISO / IEC14496-2), or MPEG4AVC standard (Moving Picture Experts Group phase 4 Advanced Video Coding, ISO14496-10) is known. Then, an image encoding apparatus that performs encoding according to these standards has been developed.
[0003]
In addition, as an image encoding device that encodes a plurality of moving images in parallel and transmits them over a network, it has a plurality of systems of encoding processing means such as frequency converters, quantizers, variable length encoders, etc. Therefore, an image encoding device that controls the output bit rate of each encoded moving image and keeps the overall bit rate constant has been proposed (see Japanese Patent Application Laid-Open No. 2003-228688).
This conventional image encoding device that encodes a plurality of moving images in parallel converts an input moving image from a spatial domain to a frequency domain with a frequency converter, and further according to a quantization coefficient with a quantizer. Quantization is performed and converted into a variable-length code by a variable-length encoder and output. A quantization coefficient is calculated according to a value obtained by summing up the complexity of the first moving image and the second moving image that are input. The output bit rate is controlled by determining and performing quantization.
[Patent Document 1] Japanese Patent Publication No. 9-512417 [Disclosure of the Invention]
[Problems to be solved by the invention]
[0004]
The above-described conventional image encoding apparatus that encodes a plurality of moving images in parallel has the same number of encoding means as the number of input moving images, so that the number of input moving images is constant. It is suitable for the form.
By the way, as a usage mode of the image encoding device, there may be a usage mode in which the number of input moving images is not always constant. For example, an image encoding device is incorporated in a television broadcast receiving device or the like, and receives several videos arbitrarily selected by the user from videos broadcast on a plurality of channels in parallel. It is assumed that the image encoding apparatus is used in such a manner that it is compressed and stored in a recording medium such as an optical disk by performing the conversion processing.
[0005]
In order to cope with a usage mode in which the number of input moving images is not always constant, the above-described conventional image encoding device previously includes encoding means corresponding to the maximum number of moving images that allow input. It will be built in. However, if the user inputs a smaller number of moving pictures than the maximum number, some encoding means are not used and are extra. This is not preferable from the viewpoint of efficient use of internal circuits constituting the image coding apparatus.
[0006]
Therefore, the present invention has been made in view of such problems, and an image encoding apparatus that performs an encoding process in response to an input of an arbitrary number of moving images by effectively utilizing a finite number of encoding means. The purpose is to provide.
[Means for Solving the Problems]
[0007]
In order to solve the above problems, an image encoding device according to the present invention is an image encoding device that performs encoding processing including motion prediction processing in order to compress a moving image, and the number of moving images to be encoded The number acquisition means for acquiring the number, the moving picture acquisition means for acquiring one or more moving pictures to be encoded, and the motion prediction processing method that affects the amount of calculation of the motion prediction processing are acquired by the number acquisition means. In accordance with the number of videos, the motion prediction processing method specifying means for specifying the calculation amount to decrease as the number increases, and the moving picture acquired by the moving picture acquisition means are subjected to encoding processing, and the acquired moving picture Encoding means for performing time-division encoding processing for each moving image if the number is a plurality, and the encoding means is specified by the motion prediction processing method specifying means as part of the encoding processing It characterized by having a motion prediction unit performing motion prediction processing by the processing method.
[0008]
Here, the calculation amount means the processing amount of the encoding process required for one moving image for the circuit responsible for the encoding process, and a large calculation amount means that it takes time to complete the process. means.
In addition, when the circuit responsible for the encoding process realizes the encoding process by a processor that executes software, the amount of calculation means the number of instructions or the instruction execution time of the arithmetic instruction for realizing the encoding process. .
【The invention's effect】
[0009]
With the above-described configuration, the image encoding device according to the present invention effectively uses the encoding means by adjusting the amount of calculation required for the encoding process according to the number of moving images to be encoded in parallel, An encoding process corresponding to input of an arbitrary number of moving images can be performed.
That is, the image encoding device according to the present invention encodes, for example, one moving image in order to reduce the amount of calculation required for motion prediction processing per moving image when the number of moving images to be encoded is large. The amount of computation required and the total amount of computation required to encode a plurality of moving images can be suppressed to a certain level. It should be noted that the motion prediction process requires a large amount of calculation in the encoding process compared to other processes in practice, and suppressing the calculation amount of the motion prediction process greatly reduces the calculation amount of the encoding process. Affect.
[0010]
Therefore, although this image encoding device cannot be said to have a possibility of deteriorating the compression rate or the image quality because of the suppression of the calculation amount, the number of moving images to be encoded without particularly increasing the device cost. Any number of images can be encoded at the same time, and there is a useful effect that encoding of all moving images can be completed within a certain amount of time. In addition, as a method for preventing the deterioration of the image quality, a method of reducing the quantization step size related to the quantization process in the encoding process can be employed.
[0011]
In addition, the motion prediction processing method specifying means associates each numerical value and each parameter for specifying the processing method of the motion prediction processing with a relationship in which the amount of calculation of the motion prediction processing decreases as the number increases. It has a storage unit that pre-stores a table, and transmits a parameter in the table corresponding to the number acquired by the number acquisition unit to the motion prediction unit, so that the processing method of the motion prediction process is In particular, the motion prediction unit may perform a motion prediction process using the specified processing method based on the parameter transmitted by the motion prediction processing method specifying unit.
[0012]
As a result, the processing method of the motion prediction process can be specified using a table that has been appropriately adjusted in advance through experiments or the like. It is possible to change the processing method of the motion prediction process according to the number of moving images to be converted. It should be noted that the table is specified so that the total amount of computation required for motion prediction processing for each moving image is within a certain range regardless of the number of moving images to be encoded. If adjusted, it is possible to appropriately encode each moving image in parallel, regardless of whether the number of moving images to be encoded is two or three.
[0013]
In addition, the motion prediction processing method specifying unit determines a processing method related to the upper limit value of the number of reference image frames in the motion prediction process as the number of reference image frames increases according to the number acquired by the number acquisition unit. The motion prediction unit is specified by the motion prediction processing method specifying means for each processing target block in each image frame to be motion prediction processed in the motion prediction processing process. The motion vector may be searched from a number of reference image frames that do not exceed the upper limit value of the number of reference image frames in the processing method.
[0014]
Note that the moving image is composed of a plurality of image frames, and the motion prediction process searches for a reference image block having a high similarity in the reference image frame with respect to each processing target block divided within each image frame. This is a process for obtaining a motion vector indicating the position of the reference image block and obtaining difference data that is a difference between the image data of the processing target block and the reference image block, and the obtained data is an encoding process subsequent to the motion prediction process. Used in the process.
[0015]
Therefore, when the number of moving images to be encoded is large, the image encoding device according to the present invention limits the number of reference image frames in which the reference block is located in the motion prediction process for each moving image to a smaller extent. This reduces the amount of computation required to search for a reference block having a high degree of similarity with each processing target block in the image frame. As a result, the finite computation processing capability of the image coding apparatus is effectively utilized, and an arbitrary number of blocks can be obtained. The moving image can be encoded at the same time.
[0016]
Further, the motion prediction processing method specifying means narrows the search range as the number of processing methods related to the motion vector search range in the motion prediction processing increases according to the number acquired by the number acquisition means. The motion prediction unit specifies the search range in the processing method specified by the motion prediction processing method specifying means in the reference image frame for each processing target block in each image frame that is a motion prediction processing target. A motion vector pointing to the reference image block may be determined by searching for a reference image block similar to the processing target block.
[0017]
As a result, if the number of moving images to be encoded is large, the search range of the reference block is narrowed in the motion prediction process for each moving image, so that the amount of calculation required for searching the reference block can be reduced. Can be encoded at the same time.
In addition, the motion prediction processing method specifying means determines the number of shape patterns as the number of processing methods related to various shape patterns of the reference image block in the motion prediction processing increases according to the number acquired by the number acquisition means. The motion prediction unit uses various shape patterns in the processing method specified by the motion prediction processing method specifying unit for each processing target block in each image frame that is a motion prediction processing target. Thus, a reference block in the reference image frame may be searched for a reference image block similar to the processing target block, and a motion vector indicating the reference image block may be determined.
[0018]
As a result, if the number of moving images to be encoded is large, the number of matching shape patterns used when searching for reference blocks is reduced in the motion prediction process for each moving image. As a result, an arbitrary number of moving images can be encoded simultaneously.
The moving image acquisition unit acquires a new moving image even during the encoding process by the encoding unit, and the number acquisition unit determines the number of moving images to be encoded during the encoding process by the encoding unit. The motion prediction processing method specifying means may specify the motion prediction processing method each time the number of moving images is acquired by the number acquisition means.
[0019]
As a result, even if the number of moving images input to the image encoding device changes, the motion prediction processing method is newly specified and the motion prediction processing is performed using that processing method accordingly. Can be encoded appropriately.
In addition, the moving image acquisition unit acquires a plurality of moving images to be encoded, and the image encoding device further receives receiving capability information indicating an arithmetic processing capability from an external device capable of executing the encoding process. And determining means for determining the number of moving images to be subjected to the encoding process requested to be executed from the external device based on the capability information; and determining by the determining means among the moving images acquired by the moving image acquiring means Transmitting means for transmitting the number of moving images to the external device, wherein the motion prediction processing method specifying means is responsive to a difference between the number acquired by the number acquiring means and the number determined by the determining means. Thus, the motion prediction processing method may be specified so that the amount of calculation decreases as the difference increases.
[0020]
In addition, the determination unit determines a processing method of motion prediction processing in the encoding processing that requests execution to the external device, together with the number of moving images that are targets of the encoding processing that requests execution based on the capability information. The transmitting unit may further transmit information indicating the processing method of the motion prediction process determined by the determining unit to the external device.
As a result, encoding of a plurality of moving images can be appropriately distributed and processed by a plurality of devices.
[0021]
Further, the image encoding device further specifies a quantization step size according to the number acquired by the number acquisition means so that the quantization step size decreases as the number increases. The encoding unit may include a quantization unit that quantizes the difference data obtained as a result of the motion prediction process with the quantization step size specified by the quantization step size specifying unit. Good.
[0022]
As a result, as the number of moving images to be encoded increases, image quality degradation that can occur by suppressing the amount of motion prediction processing for each moving image is reduced, and the quantization step size related to the quantization processing in the encoding processing is reduced. Interpolation is possible by improving the image quality.
An image encoding apparatus according to the present invention is an image encoding apparatus that performs an encoding process to compress a moving image, a number acquisition unit that acquires the number of moving images to be encoded, and an encoding target A video acquisition unit that acquires one or a plurality of videos and a processing method of an encoding process that affects the calculation amount of the encoding process are calculated as the number increases according to the number acquired by the number acquisition unit. The processing method specifying means for specifying the amount to be reduced and the processing method specified by the processing method specifying means perform the encoding process on the moving image acquired by the moving image acquisition means, and the acquired moving image If it is plural, it is provided with the encoding means which performs an encoding process by time division for each moving image.
[0023]
According to the image encoding device according to the present invention, the amount of calculation required for encoding processing for each moving image is adjusted according to the number of moving images to be encoded in parallel. Any number of moving images can be encoded at the same time.
Further, the processing method specifying means includes a table in which each numerical value and each parameter for specifying the processing method of the encoding process are associated with a relationship in which the calculation amount of the encoding process decreases as the number increases. It has a storage unit that stores in advance, and specifies the processing method of the encoding process by transmitting the parameters in the table corresponding to the number acquired by the number acquisition unit to the encoding unit. The encoding unit may perform the encoding process by the specified processing method based on the parameter transmitted by the processing method specifying unit.
[0024]
As a result, the processing method of the encoding process can be specified using a table that has been appropriately adjusted in advance through experiments or the like. The processing method of the encoding process can be changed according to the number of moving images to be converted. Note that the table is specified so that the total amount of computation required for the encoding process for each moving image is within a certain range regardless of the number of moving images to be encoded. If adjusted, it is possible to appropriately encode each moving image in parallel, regardless of whether the number of moving images to be encoded is two or three.
[0025]
In addition, the encoding unit includes an intra encoding unit that performs an intra-image frame encoding process on a part or all of image frames in a moving image to be processed as a part of the encoding process, and the processing method The specifying means reduces the number of search directions as the number increases in accordance with the number acquired by the number acquisition means for the processing method related to the direction for searching for a predicted value in intra-frame coding. And the intra coding unit includes, for each processing target block in the image frame, the processing target block of each processing target block along each of the search directions for the predicted value in the processing method specified by the processing method specifying means. It is good also as searching for the predicted value most similar to image data.
[0026]
Here, the intra-frame encoding process searches for a predicted value that is most similar to the image data of the processing target block among the predicted values calculated based on blocks in a predetermined direction within the same image frame as the processing target block. This is an intra encoding process for obtaining difference data that is the difference between the predicted value of the search result and the image data of the processing target block, and the obtained data is used in an encoding process step subsequent to the intra encoding process.
[0027]
As a result, if the number of videos to be encoded is large, in the intra encoding process for each video, the calculation target range for calculating the prediction value is reduced, so that the amount of calculation can be reduced. The moving image can be encoded at the same time.
In addition, the processing method specifying means determines whether the processing method relating to whether or not to use inter-frame coding is used between image frames if the number is less than a predetermined number according to the number acquired by the number acquisition means. If the number is greater than or equal to a predetermined number using encoding, it is specified not to use inter-frame image encoding, and the encoding means specifies that the processing method specified by the processing method specifying means is an image inter-frame encoding. If the processing method is to be used, inter-frame coding and intra-frame coding are performed on the moving image to be encoded, and the processing method specified by the processing method specifying means does not use inter-frame coding. In the case of a processing method, intra-frame encoding may be performed without performing inter-frame encoding on a moving image to be encoded.
[0028]
As a result, only the intra-frame encoding is performed when the number is large, so that the amount of calculation can be reduced and an arbitrary number of moving images can be encoded simultaneously.
Further, the encoding means includes a variable length encoding unit that performs variable length encoding processing as a part of the encoding processing, and the processing method specifying means includes the processing method related to variable length encoding, In accordance with the number acquired by the number acquisition means, if the number is less than a predetermined number, the calculation coding is used, and if the number is a predetermined number or more, it is specified not to use arithmetic coding, and the variable length code If the processing method specified by the processing method specifying unit uses arithmetic coding, the conversion unit performs variable length coding by arithmetic coding, and the processing method specified by the processing method specifying unit is If the processing method does not use arithmetic coding, variable length coding may be performed based on a reference table related to a predetermined code word without performing arithmetic coding.
[0029]
As a result, the amount of computation can be reduced because variable length coding is performed based on the table without using arithmetic coding, which is generally considered to have a large amount of computation when the number is large, and as a result, any number of videos can be encoded simultaneously. It becomes possible to become.
The moving image acquisition unit acquires a new moving image even during the encoding process by the encoding unit, and the number acquisition unit determines the number of moving images to be encoded during the encoding process by the encoding unit. And the processing method specifying means may specify the processing method of the encoding process each time the number of moving images is acquired by the number acquisition means.
[0030]
As a result, even if the number of moving images input to the image encoding device changes, the processing method of the encoding process is newly specified in accordance with the change, and the encoding process is performed according to that processing method. Can be encoded properly.
In addition, the moving image acquisition unit acquires a plurality of moving images to be encoded, and the image encoding device further receives receiving capability information indicating an arithmetic processing capability from an external device capable of executing the encoding process. And determining means for determining the number of moving images to be subjected to encoding processing to be requested to be executed by the external device based on the capability information, and determining by the determining means among the moving images acquired by the moving image acquiring means Transmission means for transmitting the number of videos to the external device, and the processing method specifying means according to a difference between the number acquired by the number acquisition means and the number determined by the determination means, The processing method of the encoding process may be specified so that the calculation amount decreases as the difference increases.
[0031]
Thereby, it becomes possible to appropriately distribute and process encoding of a plurality of moving images with a plurality of devices.
The image encoding apparatus further includes image quality information acquisition means for acquiring image quality information indicating required image quality, and the processing method specifying means acquires the processing method of encoding processing by the number acquisition means. The number may be specified in accordance with the image quality information acquired by the image quality information acquisition unit.
[0032]
For example, when the required image quality is high, the quantization step size is reduced and the request is met, so that encoding according to the required image quality is possible.
In addition, the image encoding device further includes a decoding unit that decodes a compressed moving image by sharing a part of the circuit with the encoding unit, and the processing method specifying unit is further decoded by the decoding unit. The processing method of the encoding process may be specified so that the amount of calculation decreases as the amount of moving images to be decoded increases according to the amount of moving images to be decoded.
[0033]
As a result, any number of moving images can be encoded in parallel with the decoding of the moving images.
An image encoding method according to the present invention is an image encoding method for performing an encoding process including a motion prediction process in order to compress a moving image, and the number acquiring step for acquiring the number of encoding target moving images. And a moving image acquisition step for acquiring one or a plurality of moving images to be encoded, and a motion prediction processing method that affects the calculation amount of the motion prediction processing, according to the number acquired by the number acquisition step, The motion prediction processing method specifying step that specifies the amount of calculation to decrease as the number increases, and the encoding processing is performed on the moving image acquired by the moving image acquisition step, and if there are a plurality of acquired moving images, each An encoding step that performs time-division encoding processing on a moving image, and the encoding step includes the motion prediction processing method specifying step as a part of the encoding processing. Characterized in that it comprises a motion estimation sub-step of performing motion prediction process at the specified processing method by.
[0034]
An image encoding method according to the present invention is an image encoding method for performing an encoding process to compress a moving image, the number acquiring step for acquiring the number of moving images to be encoded, and an encoding target The moving image acquisition step for acquiring one or a plurality of moving images and the processing method of the encoding process that affects the calculation amount of the encoding process are calculated as the number increases according to the number acquired by the number acquisition step. The processing method specifying step for specifying the amount to be reduced, and the processing method specified by the processing method specifying step, performing the encoding process on the moving image acquired by the moving image acquisition step, and the acquired moving image If it is plural, it includes an encoding step of performing encoding processing in a time division manner for each moving image.
[0035]
According to these image encoding methods, the amount of calculation required for the encoding process is adjusted according to the number of moving images to be encoded in parallel, so that an arbitrary number of moving images can be processed with a certain amount of calculation amount within a certain range. Can be encoded.
BEST MODE FOR CARRYING OUT THE INVENTION
<Embodiment 1>
Hereinafter, the image coding apparatus 100 according to the first embodiment of the present invention will be described.
[0036]
<Configuration>
FIG. 1 is a configuration diagram of an image encoding device 100 according to Embodiment 1 of the present invention.
As illustrated in FIG. 1, the image encoding device 100 includes a number acquisition unit 110, a moving image acquisition unit 120, a processing method identification unit 130, and an encoding unit 140.
In the figure, there is also shown a recording medium 150 such as an optical disk for accumulating data output by the image encoding apparatus 100 encoding the moving image. Here, the moving image is also used as a term indicating moving image data including an image frame representing the image content displayed at the time of reproduction.
[0037]
Here, the number acquisition unit 110 is an input interface that receives an input of the number of moving images to be encoded from the outside of the image encoding device 100 and transmits the input to the processing method specifying unit 130.
The moving image acquisition unit 120 is a moving image input interface, and has a plurality of buffer memories such as a first buffer, a second buffer,..., An Nth buffer having a capacity capable of storing a predetermined number of image frames constituting the moving image. It has a function of receiving input of one or a plurality of moving images to be encoded from the outside of the image encoding device 100 and storing each moving image in each internal buffer memory. Since the moving image is so-called stream data, when the total data amount of the moving image is large, the moving image acquisition unit 120 continues even while the image encoding device 100 is partially encoding the moving image. Is acquired.
[0038]
The processing method specifying unit 130 is realized by a memory and a processor, and includes, as functional components, a motion prediction processing method specifying unit 131, an intra coding processing method specifying unit 132, a quantization step size specifying unit 133, and a variable length coding process. A function that has a method specifying unit 134, specifies a processing method of the encoding process in the encoding unit 140 according to the number transmitted from the number acquisition unit 110, and transmits a parameter indicating the processing method to the encoding unit 140 Have
[0039]
The encoding unit 140 is realized by a memory, each circuit group that handles each part of the encoding process, and a control processor that controls each circuit group, and includes a moving image selection unit 141, a motion prediction unit 142, and an intra encoding unit 143. , A frequency conversion unit 144, a quantization unit 145, a variable length encoding unit 146, and an output unit 147, and has a function of reading a video from the buffer memory of the video acquisition unit 120, encoding it, and recording it on the recording medium 150. Here, this encoding is described as being performed in conformity with the MPEG4 AVC standard.
[0040]
The motion prediction processing method specifying unit 131 of the processing method specifying unit 130 has a motion prediction processing method table that is one area of the memory, and based on the number transmitted from the number acquisition unit 110 based on the motion prediction processing method table. Thus, the motion prediction unit 142 has a function of specifying a processing method of motion prediction processing. The motion prediction method table will be described in detail later.
The intra coding processing method specifying unit 132 has an intra coding processing method table that is an area of the memory, and based on the intra coding processing method table, according to the number transmitted from the number obtaining unit 110, the intra coding processing method table is used. The encoding unit 143 has a function of specifying a processing method for intra encoding processing. The intra coding method table will be described in detail later.
[0041]
The quantization step size specifying unit 133 has a function of calculating a quantization step size used for quantization in the quantization unit 145 by performing a predetermined calculation based on the number transmitted from the number acquisition unit 110. . This predetermined operation is a predetermined operation so that the quantization step size decreases as the number increases, and when the number is 1, it is equivalent to a conventional practical image encoding device. This is an operation determined to have a quantization step size of about.
[0042]
The variable-length encoding processing method specifying unit 134 has a function of specifying a variable-length encoding processing method based on the number transmitted from the number acquisition unit 110.
Specifically, this function uses a method of encoding based on a pre-established event / codeword comparison table assuming the statistics of events to be variable-length encoded, or further reduces the compression rate. This is a function for specifying whether to use a method called arithmetic coding in which encoding is performed while dynamically determining a code word by arithmetic operation based on a past event occurrence probability when encoding is performed in order to enhance.
[0043]
In addition, the moving image selection unit 141 of the encoding unit 140 acquires the number of moving images to be encoded via the processing method specifying unit 130, and according to the number of the moving image selection unit 141, from the buffer memory for the number of the moving image acquisition unit 120 A moving image is read cyclically every time, and each moving image is assumed to be a series of digitized image frames, and an image frame group constituting each moving image is transmitted to the motion prediction unit 142 or the intra coding unit 143. This is a circuit having a function of starting the encoding process in a time division manner for each moving image.
[0044]
In this time division processing, for example, when there are two moving images, the moving image selection unit 141 switches the moving images to be read from the buffer memory alternately for each image frame, and sequentially selects a predetermined size such as a so-called macroblock unit. This is realized by transmitting to the motion prediction unit 142 or the intra coding unit 143 in block units.
The motion prediction unit 142 can access the buffer memory of the video acquisition unit 120 for each video, receives a parameter indicating the processing method of the specified motion prediction process from the motion prediction processing method specifying unit 131, and the processing method Thus, the circuit has a function of executing a motion prediction process in units of blocks of a predetermined size with respect to a moving image frame selected by the moving image selection unit 141.
[0045]
The motion prediction process is a so-called inter-frame image encoding process. For each block in the image frame, the block (hereinafter referred to as “processing target block”) in another image frame having a high degree of similarity. A block (hereinafter referred to as a “reference block”) is searched, a motion vector indicating the position of the reference block is calculated, and difference data that is a difference value of image data between the processing target block and the reference block of the search result is calculated. This is a process of calculating and transmitting difference data to the frequency converter 144. Here, the high similarity between the image data of the processing target block and the reference block means that the distribution tendency of the image data of both blocks is similar, and the similarity is the highest when the image data of both blocks match. high.
[0046]
The intra coding unit 143 receives a parameter indicating the processing method of the specified intra coding process from the intra coding processing method specifying unit 132, and the image frame of the moving image selected by the moving image selection unit 141 according to the processing method Is a circuit having a function of executing intra coding processing in units of blocks of a predetermined size in the image frame.
Intra coding processing is so-called image frame coding processing. For each block in the image frame, the image data of the block and blocks in several predetermined directions around the same image frame This is a process of searching for a predicted value that is most similar to a predicted value based on, calculating difference data that is a difference value between the image data and the predicted value, and transmitting the difference data to the frequency converter 144. .
[0047]
The frequency transform unit 144 performs orthogonal transform on the difference data transmitted from the motion prediction unit 142 or the intra coding unit 143 so as to suppress the numerical value constituting the difference data to be small by removing the spatial redundancy. , A circuit having a function of transmitting the result of the orthogonal transformation to the quantization unit 145.
The quantization unit 145 obtains the difference data after being orthogonally transformed by the frequency conversion unit 144, and in order to suppress the numerical values constituting the difference data to be smaller, the quantization step size specifying unit 133 4 is a circuit having a function of performing quantization by rounding the result obtained by dividing by the quantization step size specified in (1) to an integer value and transmitting the quantization result to the variable-length encoding unit 146.
[0048]
The variable length coding unit 146 obtains the data obtained as a result of the quantization performed by the quantization unit 145, and obtains the data amount to further compress the data amount by expressing the information with higher appearance frequency with a shorter code. This is a circuit having a function of performing variable-length coding on data by the processing method specified by the variable-length encoding processing method specifying unit 134 and transmitting the result to the output unit 147.
The output unit 147 is a circuit having a function of recording the encoded data transmitted from the variable length encoding unit 146 on the recording medium 150 while distinguishing the data for each moving image.
[0049]
<Data>
Hereinafter, the table data stored in the processing method specifying unit 130 will be described.
FIG. 2 is a diagram illustrating a configuration and example contents of the motion prediction processing method table.
As shown in the figure, the motion prediction processing method table included in the motion prediction processing method specifying unit 131 associates the number of moving images 51 with the reference image frame number upper limit 52, the reference image block search range 53, the search pattern 54, and the like. It is a table. Note that the number 51 of moving images takes a value from 1 to a predetermined N, for example, four.
[0050]
Here, the reference image frame number upper limit 52 indicates the maximum number allowed as the number of image frames referred to by the processing target block in a certain image frame in the motion prediction process.
The reference image block search range 53 indicates the search range to which to search based on the position of the processing target block based on the range in which a reference image block having a high similarity with a certain processing target block is searched.
[0051]
The search pattern 54 indicates the size and shape of the processing target block and the reference image block. Specifically, the search pattern 54 is a pattern of 16 pixels wide × 16 pixels high, a pattern of 16 pixels wide × 8 pixels high, Pattern of width 8 pixels x height 16 pixels, pattern of width 8 pixels x height 8 pixels, pattern of width 8 pixels x height 4 pixels, pattern of width 4 pixels x height 8 pixels, width 4 pixels x height This indicates which of the 7 patterns of 4 pixels is used to search for the reference image block.
[0052]
According to the motion prediction processing method table illustrated in FIG. 2, when there is one moving image, the upper limit is two reference image frames, and the search range of the reference image block is from −15 pixels based on the processing target block. It is possible to specify a motion prediction processing method in which the range is +15 pixels and all seven patterns are applicable as a combination of the size and shape of the search pattern. When there are two moving images, the upper limit is one reference image frame, the search range of the reference image block is a range from −15 pixels to +15 pixels based on the processing target block, and the size and shape of the search pattern 4 patterns of a pattern of width 16 pixels × height 16 pixels, a pattern of width 16 pixels × height 8 pixels, a pattern of width 8 pixels × height 16 pixels, and a pattern of width 8 pixels × height 8 pixels. The processing method of the motion prediction process that can be applied can be specified.
[0053]
FIG. 3 is a diagram illustrating a configuration and an example of contents of an intra coding processing method table.
The intra coding processing method table included in the intra coding processing method specifying unit 132 is a table in which the number of moving images 61 and the prediction direction 62 are associated with each other as illustrated in FIG. Note that the number 61 of moving images takes a value from 1 to N, for example, a predetermined number.
Here, the prediction direction 62 indicates the upper limit of the direction in which a prediction value is obtained for a block in a certain image frame in the intra coding process. This direction is any one of nine directions from prediction mode 0 to prediction mode 8 for a block of 4 pixels wide × 4 pixels high.
[0054]
That is, with respect to the prediction target block, with reference to the block, prediction mode 0 for determining the value of the upper block as a prediction value, prediction mode 1 for determining the value of the left block as a prediction value, and the value of the upper block Predictive mode 2 in which the average value of the block values in the left direction and the value in the left direction is determined as the prediction value, and the prediction value is set in the 45 degree direction using the block value in the upper direction and the block value in the upper right direction for the prediction target block Prediction mode 3 to be determined, Prediction mode 4 to determine the prediction value in the 135 degree direction using the value of the block in the upward direction and the value of the block in the left direction for the prediction target block, and similarly, the prediction value in the direction of about 112 degrees Prediction mode 5 to be determined, similarly prediction mode 6 to set a prediction value in the direction of about 158 degrees, prediction mode 7 to set the prediction value in the direction of about 68 degrees, prediction to set the prediction value in the direction of about 203 degrees Prediction direction 62 of the modes 8 is selected.
[0055]
According to the intra coding processing method table illustrated in FIG. 3, nine directions of prediction modes 0 to 8 are applicable when there is one moving image, and prediction modes 0 to 8 are applicable when there are two moving images. The processing method of the intra-encoding processing that can apply the five directions of 4 can be specified.
<Operation>
Hereinafter, the operation of the image coding apparatus 100 having the above-described configuration will be described.
[0056]
The image encoding device 100 receives an input of an encoding target moving image and the number of moving images from the outside via an input interface, and the processing method specifying unit 130 specifies a processing method of encoding processing based on the number of moving images. A processing method specifying process is executed, and an encoding process for each moving image is executed by the encoding unit 140 according to the result.
FIG. 4 is a flowchart showing the processing method specifying process by the processing method specifying unit 130. Hereinafter, the operation of the processing method specifying unit 130 will be described with reference to FIG.
[0057]
The processing method specifying unit 130 first acquires the number of encoding target moving images from the number acquisition unit 110 (step S11), and the motion prediction processing method specifying unit 131 performs motion according to the number based on the motion prediction processing method table. Each parameter indicating the prediction processing method is specified (step S12), and the parameter indicating the intra encoding processing method is specified according to the number by the intra encoding processing method specifying unit 132 based on the intra encoding processing method table ( Step S13).
[0058]
Further, the processing method specifying unit 130 calculates a quantization step size by performing a predetermined calculation according to the number by the quantization step size specifying unit 133 (step S14), and the variable length coding processing method specifying unit 134. By comparing the number and a predetermined threshold value set in advance as 2 for example (step S15), if the number is less than the threshold value, a variable length coding processing method is shown so that arithmetic coding is used. A parameter is specified (step S16), and if the number is equal to or greater than a threshold, a parameter indicating a variable length coding processing method is specified so that arithmetic coding is not used (step S17). Parameters indicating the processing method, parameters indicating the intra coding processing method, quantization step size, and parameters indicating the variable length coding processing method. The meter is transmitted to the encoding unit 140 (step S18), and ends the processing method specifying process.
[0059]
By such processing method specifying processing, the processing method is specified so that the larger the number of moving images to be encoded, the smaller the processing amount of the encoding processing.
However, the quantization step size related to quantization becomes smaller as the number of moving images to be encoded increases, and as a result, the amount of quantization processing increases. Note that the amount of quantization processing, for example, the amount of computation when executed by the processor is sufficiently small in the practical range compared to the motion prediction processing, and the number of processing increases due to the processing method specifying processing described above. In this case, the amount of motion prediction processing is reduced, so that the image quality tends to deteriorate. In this case, in order to compensate for the deteriorated image quality, the quantization step size is reduced in order to reduce the image quality deterioration due to quantization. Like to do.
[0060]
FIG. 5 is a flowchart showing the encoding process by the encoding unit 140. Hereinafter, the operation of the encoding unit 140 will be described with reference to FIG.
First, according to the number transmitted from the processing method specifying unit 130, the moving image selection unit 141 of the encoding unit 140 has the number of moving images sequentially acquired from the outside by the moving image acquisition unit 120 and temporarily accumulated. From each buffer memory, moving images are switched and read sequentially for each image frame (step S21).
[0061]
Subsequently, the encoding unit 140 can determine whether the image frame of the moving image read by the moving image selection unit 141 is an image frame targeted only for the intra coding process or a target of the motion prediction process according to a predetermined criterion. It is determined whether it is an image frame (step S22), and if it is an image frame that is only subject to intra coding processing, the image is within the range of the prediction direction according to the parameters specified by the intra coding processing method specifying unit 132. The intra encoding unit 143 performs an intra encoding process by searching for an optimal prediction value for each block in the frame and obtaining difference data that is a difference between the image data of the block and the prediction value (step S23).
[0062]
If it is determined in step S22 that the image frame is not the target of only the intra coding process, the processing target block is equal to or less than the upper limit number indicated by the parameter according to the parameter specified by the motion prediction processing method specifying unit 131. The reference pattern is searched for from the search range of the reference image block in the reference image frame indicated by the parameter, and the search pattern used for searching the reference block among the processing target blocks is a parameter. The reference image block having the shape and size shown in FIG. 6 is searched for a reference image block having a high degree of similarity to the processing target block, a motion vector indicating the position of the reference image block is obtained, and the reference image block and the processing target block are obtained. To obtain difference data, which is the difference between image data and The measuring process performed by the motion prediction unit 142 (step S24).
[0063]
Note that when the reference image block is referred to from a plurality of image frames because the parameter indicating the upper limit of the number of reference image frames is 2 or more, the average value of the image data for the reference image block searched from each image frame is The difference from the image data of the processing target block is obtained as difference data.
Further, following step S23 or step S24, the frequency conversion unit 144 of the encoding unit 140 performs orthogonal transform on the obtained difference data (step S25), and the quantization unit is performed on the resultant data. 145 performs quantization with the quantization step size specified by the quantization step size specifying unit 133 (step S26).
[0064]
Following the quantization, the variable length encoding unit 146 uses the variable length code for the quantized data according to the parameters specified by the variable length encoding processing method specifying unit 134 with or without arithmetic coding. (Step S27), and the output unit 147 records the data after variable length coding on the recording medium 150 by distinguishing it for each moving image (Step S28), thereby encoding the image frame (Step S22). ˜S28) ends.
[0065]
When there are a plurality of moving images, that is, when data indicating a plurality of numbers is acquired by the number acquisition unit 110, an encoding process for a certain image frame in the first moving image is performed, and then another moving image is acquired. After the encoding process (steps S22 to S28) is performed, the encoding process is sequentially performed for the number of moving images, and the encoding process for the next image frame is started for each moving image from the first moving image. The encoding process is performed cyclically.
[0066]
When data indicating one is acquired by the number acquisition unit 110, encoding processing is sequentially performed on each image frame for one moving image. Each process (steps S22 to S28) in the encoding process illustrated in FIG. 5 is realized by pipeline processing in units of blocks.
<Modification>
Hereinafter, an image encoding device 200 that is a modification of the image encoding device 100 will be described.
[0067]
The image coding apparatus 200 is an apparatus to which the image coding apparatus 100 is added with a function of accepting an input of a requested image quality from the outside and coding while maintaining the image quality corresponding to the request.
FIG. 6 is a configuration diagram of the image encoding device 200.
The image encoding device 200 includes an image quality acquisition unit 211, a number acquisition unit 110, a moving image acquisition unit 120, a processing method identification unit 230, and an encoding unit 140, as shown in FIG. Of the constituent elements of the image encoding apparatus 200, the same constituent elements as those of the image encoding apparatus 100 are denoted by the same reference numerals in FIG. 6 as those in FIG. 1, and detailed descriptions thereof are omitted here. To do.
[0068]
Here, the image quality acquisition unit 211 is an input interface having a function of acquiring image quality information indicating image quality input by a user or the like and transmitting the acquired image quality information to the processing method specifying unit 230. The image quality information is, for example, numerical information that expresses the height of the image quality by the size of a number.
Further, the processing method specifying unit 230 includes a motion prediction processing method specifying unit 131, an intra coding processing method specifying unit 132, a quantization step size specifying unit 233, and a variable length coding processing method specifying unit 134.
[0069]
The quantization step size specifying unit 233 performs a predetermined calculation based on the number transmitted from the number acquisition unit 110 and the image quality information transmitted from the image quality acquisition unit 211, thereby performing the quantization in the quantization unit 145. It has a function of calculating the quantization step size used for the conversion. This predetermined calculation is a predetermined calculation so that the quantization step size decreases as the number increases, and the quantization step size decreases as the image quality indicated by the image quality information increases. .
[0070]
Therefore, according to the image encoding device 200, the quantization step size is specified reflecting the image quality information acquired by the image quality acquisition unit 211, and the quantization unit 145 performs quantization with the quantization step size. Therefore, encoding is performed while maintaining the requested image quality.
<Embodiment 2>
Hereinafter, an image coding system according to Embodiment 2 of the present invention will be described.
[0071]
<Configuration>
FIG. 7 is a configuration diagram of an image coding system according to Embodiment 2 of the present invention.
This image encoding system is a system in which encoding processing for a moving image is distributed by a plurality of image encoding devices, and as shown in the figure, by an image encoding device 300 and an image encoding device 400 connected to a network. Composed. Among the components of these image encoding devices, the same components as those of the image encoding device 100 shown in the first embodiment are denoted by the same reference numerals as those in FIG. 1 in FIG. Detailed description is omitted here.
[0072]
The image encoding device 300 obtains information indicating the number of moving images and the moving image from the outside, and causes the image encoding device 400 to share the encoding processing for some moving images while also encoding the other moving images. The number acquisition unit 110, the moving image acquisition unit 120, the processing method identification unit 130, the encoding unit 140, and the processing amount management unit 301 are provided.
In addition, the image encoding device 400 includes a processing method specifying unit 130, an encoding unit 140, and a processing amount management unit 402.
[0073]
The processing amount management unit 301 of the image encoding device 300 has a memory in which processing amount information indicating the processing capability of the device itself is stored in advance, and inquires the processing capability of the image encoding device 400 to obtain a response. Based on the processing capability, information indicating the number acquired from the outside by the number acquisition unit 110, and the processing capability of the own device, the number of moving image encoding processing requests to the image encoding device 400. The information indicating the number of lines determined to be determined and requested and the moving images acquired from the outside by the moving image acquiring unit 120 for the number are transmitted to the image encoding device 400, and the remaining moving images are image-coded. The encoding apparatus 300 has a function of transmitting the remaining number to the processing method specifying unit 130 in the image encoding apparatus 300.
[0074]
The processing amount management unit 402 of the image encoding device 400 has a memory in which processing amount information indicating the processing capability of the own device is stored in advance, and when a processing capability inquiry is received from the image encoding device 300, When receiving the processing amount information of its own device with reference to the memory and the information indicating the number from the image encoding device 300, the number of the information is stored in the image encoding device 400. It has a function of transmitting to the processing method specifying unit 130.
[0075]
The processing amount information indicating the processing capacity stored in the image encoding device 300 and the image encoding device 400 described above is, for example, a processing method of a predetermined encoding process used when encoding one moving image. When encoding is performed with uniform application, the number of moving images can be encoded at the same time.
<Operation>
Hereinafter, the operation of the image coding system having the above-described configuration will be described.
[0076]
When a plurality of moving images and information indicating the number of the moving images are input from the outside to the image encoding device 300, the image encoding device 300 requests the image encoding device 400 to encode some moving images. Then, a request process for encoding the remaining moving image is executed by the own device.
FIG. 8 is a flowchart showing request processing performed by the image encoding device 300. Hereinafter, the operation of the image encoding device 300 and the operation of the image encoding device 400 in response thereto will be described with reference to FIG.
[0077]
The image encoding device 300 inquires about the processing capability of the image encoding device 400 connected to the network, and receives processing amount information transmitted by the image encoding device 400 in response to this (step S31).
Subsequently, the processing amount management unit 301 of the image encoding device 300 obtains a ratio of the processing capability value indicated by the received processing amount information to the processing capability value indicated by the processing amount information of the own device, and calculates the ratio as the number. An integer closest to the value obtained by multiplying the number acquired by the acquisition unit 110 is determined as the requested number (step S32), and the requested number is transmitted to the image encoding device 400 to request encoding ( Step S33).
[0078]
As a result, the requested number is received by the image encoding device 400, and the processing method specifying unit 130 in the image encoding device 400 specifies each processing method of the encoding process according to the requested number. In addition, the encoding unit 140 in the image encoding device 400 receives the requested number of moving images from the image encoding device 300, and specifies each received moving image by the processing method specifying unit 130 in a time-sharing manner. The encoding process is executed in accordance with the processed processing method.
[0079]
Subsequent to step S <b> 33, the processing amount management unit 301 of the image encoding device 300 determines the number that is the difference between the number acquired by the number acquisition unit 110 and the requested number as a processing method specifying unit in the image encoding device 300. 130, the request process is terminated (step S34). As a result of step S34, the processing method specifying unit 130 of the image encoding device 300 specifies each processing method of the encoding process according to the number of differences. Then, according to each processing method, the encoding unit 140 of the image encoding device 300 performs encoding processing for the number of moving images corresponding to the difference.
<Supplement>
The image encoding device according to the present invention has been described above based on Embodiments 1 and 2. However, the image encoding device can be modified as follows, and the present invention is limited to the image encoding device described in the above embodiment. Of course not.
(1) Although the image encoding apparatus shown in the first and second embodiments records the encoded moving image on the recording medium 150, the recording medium 150 includes a CD-ROM (Compact Disk Read Only Memory), The optical disk is not limited to an optical disk such as a DVD (Digital Versatile Disk) or a BD (Blu-ray Disk), but may be a hard disk, a memory card, or the like. Further, the output unit 147 of the image encoding device may send each encoded moving image to the network instead of recording on the recording medium.
(2) Although the image encoding apparatus shown in the first and second embodiments performs encoding in conformity with the MPEG4 AVC standard, for example, even if it conforms to the MPEG2 video standard, MPEG4 visual standard, or the like. Well, the scope of application of the present invention is not particularly limited to a specific standard.
(3) The processing method specifying unit of the image encoding device described in the first and second embodiments includes a motion prediction processing method, an intra encoding processing method, a quantization processing method method according to a quantization step size, and The parameters that specify each of the processing methods of the variable-length encoding process are specified, but in addition to these, for example, if the number of moving images to be encoded is 3 or more, a predetermined number or more It is also possible to specify whether or not to execute the motion prediction process so that only the intra encoding process is performed without performing the motion prediction process, and the processing method of the encoding process not exemplified in the first and second embodiments May be changed according to the number of moving images to be encoded.
[0080]
Conversely, not all of the motion prediction processing method, intra encoding processing method, and variable length encoding processing method, but only a part of the processing methods are specified, and the video to be encoded is specified. The processing method specifying unit and the encoding unit may be configured so that only the encoding processing part related to a part of the processing methods varies depending on the number of the processing methods.
(4) The processing method specifying unit in the image encoding device described in the first and second embodiments includes a motion prediction method table and an intra encoding processing method table, and motion prediction in the encoding process using these tables. Each processing method for processing and intra coding processing is specified. However, a lookup table is similarly used to specify the processing method for quantization processing related to the quantization step size and the processing method for variable length coding processing. The parameters indicating each processing method may be specified by calculation based on a predetermined calculation formula.
[0081]
However, as the number of moving images to be encoded increases, the processing method is specified so that the amount of motion prediction processing, intra coding processing, and variable length coding processing for one moving image tends to decrease. The quantization step size needs to be specified to be smaller as the number of moving images is increased in order to suppress degradation in image quality to some extent by reducing this processing amount. .
(5) Although the quantization step size specifying unit 233 of the image encoding device shown in the modification of the first embodiment specifies the quantization step size based on the image quality information indicating the required image quality and the number, The quantization step size may be specified based on only the image quality information without depending on the number.
(6) The specific content of the processing method specifying unit in the image coding apparatus shown in FIGS. 2 to 4 in Embodiments 1 and 2 specifying the processing method according to the number is merely an example. It can be modified depending on the circuit configuration and processing performance of the encoding unit.
[0082]
In order to make it practical, processing is performed so that the processing speed of the encoding process executed by each circuit of the encoding unit is within a certain range regardless of the number of moving images to be encoded. In order to specify the method, it is desirable that a table or an arithmetic expression used for specifying the processing method in the processing method specifying unit is determined based on experiments or theoretical calculations.
(7) Each function unit constituting the encoding unit in the image encoding apparatus shown in the first and second embodiments is not limited to a circuit including only hardware, and the function is realized by executing software including software. In this case, the processor that executes the software may be provided for each functional unit, or a function corresponding to a plurality of functional units may be realized by one processor.
[0083]
The image encoding apparatus may be realized as part or all of LSI (Large Scale Integration), VLSI (Very Large Scale Integration), or the like, or may be realized by a plurality of LSIs or the like. You may implement | achieve by LSI etc. and other circuits.
(8) The number acquisition unit 110 of the image encoding device shown in the first embodiment acquires the number at any time even after the encoding unit 140 has started encoding processing for a moving image, The method specifying unit 130 may newly specify a processing method when the number changes, and the encoding unit 140 may perform encoding according to the result.
[0084]
As a result, it is possible to cope with a case where the number of input moving images varies.
(9) The image coding apparatus shown in the first embodiment is modified, and a circuit that can be used for decoding in the coding unit 140 is also used for decoding, and further, a circuit necessary for decoding is further used. In addition, a decoding function for decoding the compressed moving image may be provided.
[0085]
In this case, the processing method specifying unit 130 may be modified to specify the processing method so that the amount of calculation related to encoding for each moving image decreases as the amount of moving images to be decoded increases.
(10) The image encoding device 300 shown in the second embodiment may be modified to include a functional unit corresponding to the processing amount management unit 402 of the image encoding device in addition to the original configuration requirements. According to the modified image encoding device, the image encoding device 400 itself can request the image encoding device 400 to perform the encoding process, and when the other image encoding device 300 requests the encoding process, The moving image can be acquired and encoded through the moving image acquisition unit.
(11) The image coding system shown in the second embodiment is modified to exclude the processing method specifying unit 130 from the image encoding device 400, and the processing method specifying unit 130 of the image encoding device 300 is acquired by the number acquisition unit. The processing amount management unit 301 determines the number of requests to the image encoding device 400 and then determines the requested number of requests and the processing method. The processing method specified by the specifying unit 130 is transmitted to the image encoding device 400, and the encoding unit 140 of the image encoding device 400 generates the requested number of moving images according to the processing method received from the image encoding device 300. Encoding may be performed.
(12) A program for causing a processor to execute each process (see FIGS. 4 and 5) in the image encoding device can be recorded on a recording medium or distributed and distributed via various communication paths. Such recording media include IC cards, hard disks, optical disks, flexible disks, ROMs, and the like. The distributed and distributed program is used by being stored in a memory or the like that can be read by the processor, and the processor executes the program, thereby realizing the function of the image encoding device described in the first embodiment. Will come to be.
[Industrial applicability]
[0086]
The image encoding apparatus according to the present invention can be mounted and used in a hard disk recorder, a DVD recorder, or the like.
[Brief description of the drawings]
[0087]
FIG. 1 is a configuration diagram of an image encoding device 100 according to Embodiment 1 of the present invention.
FIG. 2 is a diagram showing a configuration and example contents of a motion prediction processing method table.
FIG. 3 is a diagram illustrating a configuration and an example of contents of an intra encoding processing method table.
FIG. 4 is a flowchart showing processing method specifying processing by a processing method specifying unit 130;
FIG. 5 is a flowchart showing an encoding process performed by an encoding unit 140.
6 is a configuration diagram of an image encoding device 200. FIG.
FIG. 7 is a configuration diagram of an image coding system according to Embodiment 2 of the present invention.
FIG. 8 is a flowchart showing a request process performed by the image encoding device 300;
[Explanation of symbols]
[0088]
100, 200, 300, 400 Image encoding device 110 Number acquisition unit 120 Movie acquisition unit 130, 230 Processing method specifying unit 131 Motion prediction processing method specifying unit 132 Intra coding processing method specifying unit 133, 233 Quantization step size specifying unit 134 Variable Length Coding Processing Method Identification Unit 140 Encoding Unit 141 Video Selection Unit 142 Motion Prediction Unit 143 Intra Coding Unit 144 Frequency Conversion Unit 145 Quantization Unit 146 Variable Length Coding Unit 147 Output Unit 150 Recording Medium 211 Image Quality Acquisition Unit 301, 402 Processing amount management unit

Claims (22)

An image encoding device that performs an encoding process including a motion prediction process to compress a moving image,
Number acquisition means for acquiring the number of videos to be encoded;
Moving image acquisition means for acquiring one or more moving images to be encoded;
Motion prediction processing method specification that specifies the processing method of motion prediction processing that affects the calculation amount of motion prediction processing so that the calculation amount decreases as the number increases according to the number acquired by the number acquisition unit Means,
An encoding unit that performs encoding processing on the moving image acquired by the moving image acquisition unit, and performs encoding processing on a time-division basis for each moving image if there are a plurality of acquired moving images;
The image encoding device, wherein the encoding unit includes a motion prediction unit that performs a motion prediction process using the processing method specified by the motion prediction processing method specifying unit as part of the encoding process. The motion prediction processing method specifying means includes a table in which each numerical value and each parameter for specifying the motion prediction processing processing method are associated with a relationship in which the amount of calculation of the motion prediction processing decreases as the number increases. It has a storage unit that stores in advance, and specifies the processing method of the motion prediction process by transmitting the parameters in the table corresponding to the number acquired by the number acquisition means to the motion prediction unit. ,
The image coding apparatus according to claim 1, wherein the motion prediction unit performs a motion prediction process using the specified processing method based on the parameter transmitted by the motion prediction processing method specifying unit. The motion prediction processing method specifying means determines the processing method related to the upper limit value of the number of reference image frames in the motion prediction processing according to the number acquired by the number acquisition means as the number of reference image frames increases. Specify a lower limit,
The motion prediction unit, in a motion prediction process, for each processing target block in each motion prediction processing target image frame, an upper limit value of the number of reference image frames in the processing method specified by the motion prediction processing method specifying means The image encoding apparatus according to claim 1, wherein motion vectors are searched from a number of reference image frames not exceeding. The motion prediction processing method specifying means sets a processing method related to a motion vector search range in motion prediction processing so that the search range becomes narrower as the number increases according to the number acquired by the number acquisition means. Identify,
The motion prediction unit, for each processing target block in each image frame subject to motion prediction processing, from within a search range in the processing method specified by the motion prediction processing method specifying means in the reference image frame, The image encoding device according to claim 1, wherein a reference image block similar to the processing target block is searched to determine a motion vector indicating the reference image block. The motion prediction processing method specifying means is a processing method related to various shape patterns of reference image blocks in the motion prediction processing, and the number of shape patterns decreases as the number increases according to the number acquired by the number acquisition means. To be specific,
The motion prediction unit uses each shape pattern in the processing method specified by the motion prediction processing method specifying means for each processing target block in each image frame to be motion prediction processing, to reference in a reference image frame The image coding apparatus according to claim 1, wherein a block is searched for a reference image block similar to the processing target block to determine a motion vector indicating the reference image block. The moving image acquisition means acquires a new moving image even during the encoding process by the encoding means,
The number acquisition unit newly acquires the number of moving images to be encoded during the encoding process by the encoding unit,
The image coding apparatus according to claim 1, wherein the motion prediction processing method specifying unit specifies a processing method of motion prediction processing every time the number of moving images is acquired by the number acquisition unit. The moving image acquisition means acquires a plurality of moving images to be encoded,
The image encoding device further includes:
Receiving means for receiving capability information indicating an arithmetic processing capability from an external device capable of executing encoding processing;
A determination unit that determines the number of moving images to be subjected to an encoding process that is requested to be executed by the external device based on the capability information;
Transmission means for transmitting, to the external device, the number of videos determined by the determination means among the videos acquired by the video acquisition means,
The motion prediction processing method specifying unit is configured to perform motion prediction processing so that the amount of calculation decreases as the difference increases according to the difference between the number acquired by the number acquisition unit and the number determined by the determination unit. The image encoding apparatus according to claim 1, wherein the processing method is specified. The determining means determines, based on the capability information, a processing method of motion prediction processing in an encoding process that requests execution to the external device, together with the number of moving images that are targets of the encoding process that requests execution,
The image encoding device according to claim 7, wherein the transmission unit further transmits information indicating a processing method of motion prediction processing determined by the determination unit to the external device. The image encoding apparatus further includes a quantization step size specifying unit that specifies a quantization step size so that the quantization step size decreases as the number increases according to the number acquired by the number acquisition unit. With
The said encoding means has a quantization part which quantizes the difference data obtained as a result of a motion estimation process by the quantization step size specified by the said quantization step size specification means. The image encoding device described. An image encoding device that performs an encoding process to compress a moving image,
Number acquisition means for acquiring the number of videos to be encoded;
Moving image acquisition means for acquiring one or more moving images to be encoded;
A processing method specifying means for specifying the processing method of the encoding process that affects the calculation amount of the encoding process according to the number acquired by the number acquisition means so that the calculation amount decreases as the number increases; ,
Encodes the moving image acquired by the moving image acquisition unit with the processing method specified by the processing method specifying means, and if there are multiple acquired moving images, encodes each moving image as a target in a time division manner An image encoding apparatus comprising: encoding means for performing processing. The processing method specifying unit stores in advance a table in which each numerical value and each parameter for specifying the processing method of the encoding process are associated with a relationship in which the calculation amount of the encoding process decreases as the number increases. Identifying the processing method of the encoding process by transmitting the parameters in the table corresponding to the number acquired by the number acquisition means to the encoding means,
The image coding apparatus according to claim 10, wherein the encoding unit performs an encoding process using the specified processing method based on the parameter transmitted by the processing method specifying unit. The encoding unit includes an intra encoding unit that performs an intra-image frame encoding process on a part or all of image frames in a moving image to be processed as a part of the encoding process,
The processing method specifying means determines the number of directions to be searched as the number increases according to the number acquired by the number acquisition means according to the number acquired by the number acquisition means. Specific to be small,
The intra coding unit is configured to perform processing with respect to each processing target block in the image frame along with the image data of the processing target block along each of the search directions for the predicted value in the processing method specified by the processing method specifying unit The image encoding apparatus according to claim 10, wherein a similar predicted value is searched. The processing method specifying means determines whether the processing method relating to whether or not to use inter-frame image coding is the inter-image frame coding if the number is less than a predetermined number according to the number obtained by the number obtaining means. If the number is greater than or equal to a predetermined number, specify that image interframe coding is not used,
The encoding means includes
If the processing method specified by the processing method specifying means is a processing method that uses inter-frame coding, it performs inter-frame coding and intra-frame coding on the moving image to be coded,
If the processing method specified by the processing method specifying means is a processing method that does not use inter-frame encoding, intra-frame encoding is performed without performing inter-frame encoding on the moving image to be encoded. The image encoding device according to claim 10. The encoding means includes a variable length encoding unit that performs a variable length encoding process as part of the encoding process,
The processing method specifying unit uses a calculation encoding method if the number is less than a predetermined number according to the number acquired by the number acquisition unit, and the number of the processing method relating to variable length encoding is equal to or greater than the predetermined number. If so, specify not to use arithmetic coding,
The variable length encoding unit includes:
If the processing method specified by the processing method specifying means is a processing method using arithmetic coding, variable length coding is performed by arithmetic coding,
If the processing method specified by the processing method specifying means does not use arithmetic coding, arithmetic coding is not performed, and variable length coding is performed based on a reference table related to a predetermined code word. The image encoding device according to claim 10, wherein the image encoding device is performed. The moving image acquisition means acquires a new moving image even during the encoding process by the encoding means,
The number acquisition unit newly acquires the number of moving images to be encoded during the encoding process by the encoding unit,
The image coding apparatus according to claim 10, wherein the processing method specifying unit specifies a processing method of encoding processing every time the number of moving images is acquired by the number acquisition unit. The moving image acquisition means acquires a plurality of moving images to be encoded,
The image encoding device further includes:
Receiving means for receiving capability information indicating an arithmetic processing capability from an external device capable of executing encoding processing;
A determination unit that determines the number of moving images to be subjected to an encoding process that is requested to be executed by the external device based on the capability information;
Transmission means for transmitting, to the external device, the number of videos determined by the determination means among the videos acquired by the video acquisition means,
The processing method specifying unit is configured to perform encoding processing so that the amount of calculation decreases as the difference increases according to the difference between the number acquired by the number acquisition unit and the number determined by the determination unit. The image coding apparatus according to claim 10, wherein a method is specified. The image encoding device further includes image quality information acquisition means for acquiring image quality information indicating the required image quality,
11. The processing method specifying unit specifies a processing method of encoding processing according to the number obtained by the number acquisition unit and the image quality information acquired by the image quality information acquisition unit. Image coding apparatus. The image encoding device further includes decoding means for decoding a compressed moving image by sharing a part of the circuit with the encoding means,
The processing method specifying means further specifies the processing method of the encoding process so that the amount of calculation decreases as the amount of moving images to be decoded increases according to the amount of moving images decoded by the decoding means. The image encoding device according to claim 10. An image encoding method for performing an encoding process including a motion prediction process in order to compress a moving image,
A number acquisition step for acquiring the number of videos to be encoded;
A video acquisition step of acquiring one or more videos to be encoded;
Motion prediction processing method specification that specifies the processing method of motion prediction processing that affects the calculation amount of motion prediction processing so that the amount of calculation decreases as the number increases according to the number acquired by the number acquisition step Steps,
An encoding step for performing the encoding process on the moving image acquired by the moving image acquisition step, and performing an encoding process on a time division basis for each moving image if there are a plurality of acquired moving images,
The image encoding method, wherein the encoding step includes a motion prediction sub-step for performing a motion prediction process using the processing method specified by the motion prediction processing method specifying step as a part of the encoding process. An image encoding method for performing an encoding process to compress a moving image,
A number acquisition step for acquiring the number of videos to be encoded;
A video acquisition step of acquiring one or more videos to be encoded;
A processing method specifying step for specifying the processing method of the encoding process that affects the calculation amount of the encoding process according to the number acquired by the number acquisition step so that the calculation amount decreases as the number increases. ,
In the processing method specified in the processing method specifying step, encoding processing is performed on the moving image acquired in the moving image acquisition step. If there are a plurality of acquired moving images, encoding is performed on each moving image in a time-sharing manner. An image encoding method comprising: an encoding step for performing processing. A control program for causing an apparatus capable of executing a program to execute an encoding control process to compress a moving image,
The encoding control process includes:
A number acquisition step for acquiring the number of videos to be encoded;
A video acquisition step of acquiring one or more videos to be encoded;
Motion prediction processing method specification that specifies the processing method of motion prediction processing that affects the calculation amount of motion prediction processing so that the amount of calculation decreases as the number increases according to the number acquired by the number acquisition step Steps,
An encoding step for performing the encoding process on the moving image acquired by the moving image acquisition step, and performing an encoding process on a time division basis for each moving image if there are a plurality of acquired moving images,
The encoding program includes a motion prediction sub-step for performing a motion prediction process using the processing method specified by the motion prediction processing method specifying step as a part of the encoding process. A control program for causing an apparatus capable of executing a program to execute an encoding control process to compress a moving image,
The encoding control process includes:
A number acquisition step for acquiring the number of videos to be encoded;
A video acquisition step of acquiring one or more videos to be encoded;
A processing method specifying step for specifying the processing method of the encoding process that affects the calculation amount of the encoding process according to the number acquired by the number acquisition step so that the calculation amount decreases as the number increases. ,
In the processing method specified in the processing method specifying step, encoding processing is performed on the moving image acquired in the moving image acquisition step. If there are a plurality of acquired moving images, encoding is performed on each moving image in a time-sharing manner. A control program comprising: an encoding step for performing processing.

Images